Polycapillary x-ray optics are bundles of micron size hollow tubes, inside of which x rays are propagated by total reflection much like visible light in solid fiber optics. The small critical angle for total reflection from the glass walls of the tubes, 0.06° at 27 keV, results in very high angular selectivity. The field of view of each capillary tube is limited by this angular acceptance to less than 50 microns at a source-to-optic distance of 2 cm. Each adjacent tube works in parallel so that a large area can be covered at this resolution with much higher count rate than for a single collimator. Measurements have been performed using ¹²⁵I brachytherapy seeds in Lucite phantoms using the optics and imaging detectors. Measured resolutions were detector-limited at better than 0.1 mm. Calculations for expected sensitivity and signal-to-background ratios were developed from geometrical models and show good agreement with measurements. Results indicate that the optics provide superior signal count rates to conventional collimators for geometries such as small animal imaging in which sub millimeter resolution with inch-wide or larger fields of view are desirable.

A novel jet-printing approach to fabricate thin film transistor (TFT), active matrix backplanes for x-ray imagers is described. The technique eliminates the use of photolithography and has the potential to greatly reduce the array manufacturing cost. We show how jet-printing is used to pattern the layers of the active matrix array and also to deposit semiconductor material. The technique is applied to both amorphous silicon and polymer transistors, and small prototype arrays have been fabricated and tested, including arrays with a high fill factor amorphous silicon p-i-n photodiode layer for indirect detection x-ray imaging applications. The TFT characteristics are excellent, and acquired x-ray images will be presented and compared to those from conventional TFT arrays. The printing process has been extended to flexible substrates which are important for rugged x-ray imagers, using a low temperature amorphous silicon process to accommodate the plastic substrate. Polymer TFT arrays made with jet-printed polymer solutions have also been demonstrated and we present data from arrays, and discuss options for integrating organic photodiodes or direct detection sensors. The opportunities and challenges of using polymer semiconductors in x-ray imaging arrays, are discussed and we show that the TFT performance meets the needs of radiographic imaging, although the radiation hardness and long term degradation are not sufficiently studied.

The numerous merits of x-ray imagers based on active matrix, flat-panel array technology have led to their introduction in a wide variety of x-ray imaging applications. However, under certain conditions, the performance of direct and indirect detection AMFPIs is significantly limited by the relatively modest ratio of singal to noise provided by conventional systems. While substantial reduction in the additive noise of such systems is difficult, significant enhancement of signal can be achieved through the incorporation of an amplification circuit in each pixel. In addition, innovative photodiode structures can be incorporated into indirect detection designs to enhance optical signal collection efficiency. In this paper, an investigation of these strategies, involving the design, fabrication and performance evaluation of a variety of novel, small area, indirect detection arrays, is described. Each prototype array incorporates innovative features, such as continuous photodiodes and single-stage and dual-stage in-pixel amplifiers, that are designed to provide insight into promising avenues for achieving significant singal-to-noise enhancement. This information will assist in the realization of a next generation of highly-optimized AMFPI pixel architectures whose DQE performance will be limited only by x-ray noise and x-ray converter properties under a very wide range of conditions. In this paper, the design and operation of the present prototype arrays are described and initial performance results are reported. In addition, the benefits of significant improvements to the signal-to-noise properties of AMFPIs are illustrated through cascaded systems calculations of the DQE performance of hypothetical systems.

X-ray dark-field imaging (DFI) due to refraction is under development with intension of its clinical application. In this system we have adopted an asymmetric-cut monochro-collimator (M) and an angular analyzer (A) of Si 440 diffraction at 35 keV of X-rays. By choosing an appropriate thickness T of A that satisfies the condition T = ΛN where Λ is the extinction distance and N integer the transmissivity in the region of |W| (angular parameter) < 1 should be theoretically almost zero and |W| > 1 should be approximately 70-80%. This has been experimentally proven. Under this condition the X-rays whose propagation direction may not change such as those receiving only absorption will not go into the forward diffraction direction after A but go into the diffraction direction, while the X-rays refracted by object may go into the forward diffraction direction after A. We have settled two targets of clinical views: soft tissues at joints and early check of breast cancer. A first clear image of articular cartilage of small joint was successfully obtained using a proximal interphalangeal joint that was amputated from a cadaver. Since larger view field is needed for clinical use the size of approximately 90 mm in square has been successfully achieved. Using this beam articular cartilage of knee and shoulder joints from the same cadaver have been successfully visualized. Further visibility test by the DFI is under way for a phantom of breast cancer, paraffin fixed sliced breast samples containing micro-calcification, tumor and excised breast tissue.

In image-guided procedures high-contrast objects often appear in the imaging field-of-view for the purpose of guiding treatment (e.g., markers intended to localize the target) or delivering treatment (e.g., surgical tools, or in the case of brachytherapy, radioactive seeds). In cone-beam CT reconstructions, these high-contrast objects cause severe streak artifacts, CT number inaccuracy and loss of soft-tissue visibility. We have developed an iterative approach by which high-contrast objects are localized in the 2-D projection set by re-projecting conspicuities from the first-pass 3-D reconstruction. The projection operator, which finds the unique mapping from the world coordinate system to the detector coordinate system for each view angle, is computed from a geometric calibration of the system. In each projection, a two-dimensional 2^nd order Taylor series is used to interpolate over the high-contrast objects. The interpolated surface is further modified using a local noise estimate to completely mask the objects. The algorithm has been applied to remove artifacts resulting from a small number of gold fiducial markers in patients being imaged daily with cone-beam CT for guidance of prostate radiotherapy. The algorithm has also been applied to post-operative images of a prostate brachytherapy patient in which the number of seeds can exceed ~100. In each case, the method provides excellent attenuation of image artifact and restoration of soft-tissue visibility. Using a local voxel based metric it was shown that the 2^nd order Taylor series with added noise performed best at removing the high-contrast objects from the reconstruction volume.

One big advantage in terms of image quality of modern flat panel detector systems compared to CR systems beside the better DQE of these systems is the possibility to correct for inhomogeneities of the X-ray beam and the detector (flat field correction) as well as for bad pixels. However, the used correction methods are taking a lot of time or do not cover all possible combinations of radiation quality and exposure used for patient imaging. A method is presented to achieve these correction images very easily by using a proposed method for comparing two images. This method, which has so far been used for certain noise measurements and in some cases noise reduction, can also be used for separating correlated from uncorrelated noise by correlating in frequency sub-bands the information of two images. In this study it is proven, that the uncorrelated noise image of two expositions is very similar to the correction image gained just before the two exposures. That allows to calibrate a detector quite more often and for much more beam qualities/exposures than before to achieve a better correction and another possibility of constancy testing for flat panel detectors, because the proposed method is so sensitive that it will detect single pixel changes within the detector.

Reconstruction algorithms for volumetric CT have been the focus of many studies. Several exact and approximate reconstruction algorithms have been proposed for step-and-shoot and helical scanning trajectories to combat cone beam related artifacts. In this paper, we present a closed form cone beam reconstruction formula for tilted gantry data acquisition. Although several algorithms were proposed to compensate for errors induced by the gantry tilt, none of the algorithms addresses the case in which the cone beam geometry is first rebinned to a set of parallel beams prior to the filtered backprojection. Because of the rebinning process, the amount of iso-center adjustment depends not only on the projection angle and tilt angle, but also on the reconstructed pixel location. The proposed algorithm has been tested extensively on both 16 and 64 slice VCT with phantoms and clinical data. The efficacy of the algorithm is clearly demonstrated by the experiments.

Imaging of small high-density structures, such as calcifications, with computed tomography (CT) is limited by the spatial resolution of the system. Blur causes small calcifications to be imaged with lower contrast and overestimated volume, thereby hampering the analysis of vessels. The aim of this work is to reduce the blur of calcifications by applying three-dimensional (3D) deconvolution. Unfortunately, the high-frequency amplification of the deconvolution produces edge-related ring artifacts and enhances noise and original artifacts, which degrades the imaging of low-density structures. A method, referred to as Histogram-based Selective Deblurring (HiSD), was implemented to avoid these negative effects. HiSD uses the histogram information to generate a restored image in which the low-intensity voxel information of the observed image is combined with the high-intensity voxel information of the deconvolved image. To evaluate HiSD we scanned four in-vitro atherosclerotic plaques of carotid arteries with a multislice spiral CT and with a microfocus CT (μCT), used as reference. Restored images were generated from the observed images, and qualitatively and quantitatively compared with their corresponding μCT images. Transverse views and maximum-intensity projections of restored images show the decrease of blur of the calcifications in 3D. Measurements of the areas of 27 calcifications and total volumes of calcification of 4 plaques show that the overestimation of calcification was smaller for restored images (mean-error: 90% for area; 92% for volume) than for observed images (143%; 213%, respectively). The qualitative and quantitative analyses show that the imaging of calcifications in CT can be improved considerably by applying HiSD.

Correction for non-uniform attenuation in SPECT generally requires measurements of radiation transmittance through the patient and reconstruction of the data to form an attenuation image, or mu-map. For nuclear cardiac studies it useful if the emission and transmission data for each projection view can be acquired simultaneously using non-overlapping energy windows. This simplifies the registration of the emission and transmission data. Large area transmission sources are desirable to avoid data truncation; however, 2D-planar liquid sources are cumbersome and extended solid area sources of Gd-153 or Am-247 are impractical. Co-57 sheet sources present spectral overlap problems for imaging of Tc-99m tracers. With Gd-153 line arrays, one can achieve the benefits of 2D-planar sources, low truncation and simultaneous emission/transmission measurements, using lightweight static mechanical attachments to the SPECT camera system. A new method is proposed to determine optimal positions for the lines of the transmission array based on maximizing the entropy of the transmitted flux through the patient. Transmission reconstruction using parallel beam filtered back-projection yields attenuation maps with poor spatial resolution and significant aliasing effects. The degradations of image quality become worse as the angular separations of the lines as seen by the detector increase. To improve the reconstruction of line array transmission data a maximum likelihood modified gradient algorithm was derived. The algorithm takes into account emission-to-transmission down scatter as well as the overlapping of radiation patterns of the individual lines. Ordered subset versions of algorithms are explored. Image quality is assessed with simulations based on an attenuation map derived from CT.

In the present study the applicability of x-ray fluorescence tomography for in-vivo medical imaging was investigated with respect to signal strength, background distribution and minimum detectable concentration. Tomographic imaging of the concentration distribution of suitable marker substances by the detection of the x-ray fluorescence emitted upon external excitation with x-rays has been demonstrated by other groups. However, most of these studies work with parameters that are not realistic for the medical in-vivo imaging of marker substances based on this principle; e.g. they use very small phantoms or gaseous markers. The investigated scenario uses the irradiation during a transmission computed tomography (CT) scan for the external activation of a suitable type and concentration of an x-ray fluorescence marker administered to the patient. During the irradiation, collimated and energy-resolving detectors acquire fluorescence radiation signals emitted along lines through the patient. By tomographic reconstruction of the fluorescence signal data-set, a concentration map of the marker is generated. This fluorescence image will be inherently co-registered with the high-resolution transmission CT image and can show functional or metabolic processes as an additional channel of information. The present study is based both on phantom experiments in a dedicated measurement set-up and on simulations, using various marker substances and detection concepts. Special focus was given to background reduction strategies. Moreover, the background signal in the spectral detection windows that limits the concentration resolution of the method was quantified. Signal-to-background ratios and minimum detectable marker concentrations for different scanner concepts will be presented.

Multidetector CT (MDCT) systems offer larger coverage and wider z-axis beams, resulting in larger cone angles. One impact on radiation dose is that while radiation profiles at isocenter are constant when contiguous axial scans are performed, the increased beam divergence from the larger cone angle results in significant surface dose variation. The purpose of this work was to measure the magnitude of this effect. Contiguous axial scans were acquired using an MDCT for two sizes of cylindrical phantoms and an anthropomorphic phantom. Film dosimetry and/or radiation detector measurements were performed on the surface of each phantom. Detailed mathematical models were developed for the MDCT scanner and all phantoms. Monte Carlo simulations of contiguous axial scans were performed for each phantom model. From cylindrical phantoms, film dosimetry at the surface showed differences between peak and valley that reached 50%. From the anthropomorphic phantom, measured values ranged from 7.9 to 16.2 mGy at the phantom surface. Monte Carlo simulations demonstrated these variations in both cylindrical and anthropomorphic phantoms. The magnitude of variation was also related to object size. Even when contiguous axial scans are performed on MDCT, surface radiation profiles show considerable variation. This variation will increase as MDCT cone angles increase and when non-contiguous scans (e.g. pitch > 1) are acquired. The variation is also a function of object size. While average surface doses may remain constant, peak doses may increase, which may be significant for radiation sensitive organs at or near the surface (e.g. breast, thyroid).

Conventional X-ray radiographic systems rely on transmitted photons for the production of images. Backscatter imaging makes use of the more abundant scattered photons for image formation. Specifically, incoherently (Compton) scattered X-ray photons are detected and used for image formation in this modality of medical imaging. However, additional information is obtained when the transmitted X-ray photons are also detected and used. Transmission radiography produces a two-dimensional image of a three dimensional system, therefore image information from a shallower object is often contaminated by image information from underlying objects. Backscattered x-ray imaging largely overcomes this deficiency by imaging depth selectively, which reduces corruption of shallow imaging information by information from deeper objects lying under it. Backscattered x-ray imaging may be particularly useful for examining anatomical structures at shallow depths beneath the skin. Some typical applications for such imaging might be breast imaging, middle ear imaging, imaging of skin melanomas, etc. Previous investigations, by way of theoretical calculations and computational simulations into the feasibility of this kind of imaging have uncovered high-contrast and SNR parameters. Simulations indicate that this method can be used for imaging relatively high-density objects at depths of up to approximately five centimeters below the surface. This paper presents both theoretical and experimental SNR results on this new medical imaging modality.

Healthy tissues and tumors exhibit different optical characteristics in blood volume and oxygen sufficiency. Tumor physiology is effectively monitored by non-invasively observing the changes in oxyhemoglobin and deoxyhemoglobin concentration in tissue. In this paper, we present a practical method for quantitative assessment of hemoglobin concentration and blood oxygenation based on the diffusion theory and finite element analysis. The method incorporates prior knowledge on permissible target region, and reduced the reconstruction of chromosphere concentration to an optimization procedure with simple constrain. A numerical simulation study has been conducted by using a heterogeneous phantom. The numerical results show that the reconstruction method has been successfully applied for the reconstruction of the variation of HbO₂ and HbR concentration in numerical simulation experiments.

Colon cancer alters the macroarchitecture of the colon tissue. Common changes include angiogenesis and the distortion of the tissue collagen matrix. Such changes affect the colon colouration. This paper presents the principles of a novel optical imaging method capable of extracting parameters depicting histological quantities of the colon. The method is based on a computational, physics-based model of light interaction with tissue. The colon structure is represented by three layers: mucosa, submucosa and muscle layer. Optical properties of the layers are defined by molar concentration and absorption coefficients of haemoglobins; the size and density of collagen fibres; the thickness of the layer and the refractive indexes of collagen and the medium. Using the entire histologically plausible ranges for these parameters, a cross-reference is created computationally between the histological quantities and the associated spectra. The output of the model was compared to experimental data acquired in vivo from 57 histologically confirmed normal and abnormal tissue samples and histological parameters were extracted. The model produced spectra which match well the measured data, with the corresponding spectral parameters being well within histologically plausible ranges. Parameters extracted for the abnormal spectra showed the increase in blood volume fraction and changes in collagen pattern characteristic of the colon cancer. The spectra extracted from multi-spectral images of ex-vivo colon including adenocarcinoma show the characteristic features associated with normal and abnormal colon tissue. These findings suggest that it should be possible to compute histological quantities for the colon from the multi-spectral images.

The limitations of conventional dosimeters restrict the comprehensiveness of verification that can be performed for advanced radiation treatments presenting an immediate and substantial problem for clinics attempting to implement these techniques. In essence, the rapid advances in the technology of radiation delivery have not been paralleled by corresponding advances in the ability to verify these treatments. Optical-CT gel-dosimetry is a relatively new technique with potential to address this imbalance by providing high resolution 3D dose maps in polymer and radiochromic gel dosimeters. We have constructed a 1^st generation optical-CT scanner capable of high resolution 3D dosimetry and applied it to a number of simple and increasingly complex dose distributions including intensity-modulated-radiation-therapy (IMRT). Prior to application to IMRT, the robustness of optical-CT gel dosimetry was investigated on geometry and variable attenuation phantoms. Physical techniques and image processing methods were developed to minimize deleterious effects of refraction, reflection, and scattered laser light. Here we present results of investigations into achieving accurate high-resolution 3D dosimetry with optical-CT, and show clinical examples of 3D IMRT dosimetry verification. In conclusion, optical-CT gel dosimetry can provide high resolution 3D dose maps that greatly facilitate comprehensive verification of complex 3D radiation treatments. Good agreement was observed at high dose levels (>50%) between planned and measured dose distributions. Some systematic discrepancies were observed however (rms discrepancy 3% at high dose levels) indicating further work is required to eliminate confounding factors presently compromising the accuracy of optical-CT 3D gel-dosimetry.

The purpose of this study is to perform a preliminary evaluation of a newly constructed flat panel detector (FPD)-based system for cone beam CT imaging applications. A prototype flat-panel detector-based cone beam CT imaging system has been designed and constructed by modifying a GE HiSpeed Advantage CT scanner. The prototype consists of a modified GE CT HiSpeed Advantage CT gantry, an x-ray tube, a 397mm x 298mm Varian PaxScan 4030CB real time flat-panel detector mounted on the gantry, a CT table and an on-gantry PC to control image acquisition. Another PC workstation serving as an operating console controls the CT gantry and CT table, and sends trigger pulses to dedicated electronic interface modules to control radiographic exposure and to initiate data acquisition. Captured image data sets are first stored in the on-gantry computer and then downloaded from the on-gantry PC to the operating console for high-speed 3D image reconstruction. During data acquisition, the x-ray tube and the FPD can be rotated on the gantry over Nx360 degrees due to integrated slip ring technology. With a single scan, this device is able to acquire up to three hundred two-dimensional projections (1024 x 768 x 16 bits) for direct 3D reconstruction within 10 seconds. This system was used for a series of preliminary phantom studies and small animal studies. Using the continuous scan mode of the scanner, a few hundred projections were acquired for all volume scans. Direct 3D reconstructions were obtained to evaluate the system for cone beam CT for lung imaging applications. The preliminary results indicate that the newly built flat panel detector-based cone beam CT scanner works as expected and the low contrast resolution of the FPD-based CBCT system is approaching that of a multi-slice CT.

Recent developments in two-dimensional x-ray detector technology have made volumetric Cone Beam CT (CBCT) a feasible approach for integration with conventional medical linear accelerators. The requirements of a robust image guidance system for radiation therapy include the challenging combination of soft tissue sensitivity with clinically reasonable doses. The low contrast objects may not be perceptible with MV energies due to the relatively poor signal to noise ratio (SNR) performance. We have developed an imaging system that is optimized for MV and can acquire Megavoltage CBCT images containing soft tissue contrast using a 6MV x-ray beam. This system is capable of resolving relative electron density as low as 1% with clinically acceptable radiation doses. There are many factors such as image noise, x-ray scatter, improper calibration and acquisitions that have a profound effect on the imaging performance of CBCT and in this study attempts were made to optimize these factors in order to maximize the SNR. A QC-3V phantom was used to determine the contrast to noise ratio (CNR) and f₅₀ of a single 2-D projection. The computed f₅₀ was 0.43 lp/mm and the CNR for a radiation dose of 0.02cGy was 43. Clinical Megavoltage CBCT images acquired with this system demonstrate that anatomical structures such as the prostate in a relatively large size patient are visible using radiation doses in range of 6 to 8cGy.

This work investigates the modulation transfer function (MTF) of a prototype table-top inverse-geometry volumetric CT (IGCT) system. The IGCT system has been proposed to acquire sufficient volumetric data in one circular rotation using a large-area scanned source and a narrower array of fast detectors. The source and detector arrays have the same axial, or slice, extent, thus providing sufficient volumetric coverage. A prototype system has been built using a NexRay Scanning-Beam Digital X-ray system (NexRay, Inc., Los Gatos, CA) with the C-arm gantry in the horizontal position and a stage placed between the source and detector to rotate the scanned object. The resulting system has a 16-cm in-plane field of view (FOV) and 5-cm axial FOV. Two phantoms were constructed for measuring the MTF. A 76 micron tungsten wire placed axially in a plastic frame was used to measure the in-plane MTF, and the same wire slanted at 45 degrees was used to test the isotropy of the MTF. The data were calibrated for flat-field intensity and geometric misalignment and reconstructed using a modified 3D PET algorithm. For both phantoms, slices perpendicular to the wires were reconstructed. Simulations which model the IGCT system were used to verify the MTF measurement, along with analytical predictions. The measured MTF curve was similar in shape to the predicted curve with a 10% point at 20 lp/cm compared to a predicted 18 lp/cm. Future work will also study the uniformity of the MTF across the FOV and further characterize the IGCT system.

We developed and evaluated a prototype flat-panel detector based Volume CT (VCT) scanner. We focused on improving the image quality using different detector settings and reducing x-ray scatter intensities. For the presented results we used a Varian 4030CB flat-panel detector mounted in a multislice CT-gantry (Siemens Medical Systems). The scatter intensities may severely impair image quality in flat-panel detector CT systems. To reduce the impact of scatter we tested bowtie shaped filters, anti-scatter grids and post-processing correction algorithms. We evaluated the improvement of image quality by each method and also by a combination of the several methods. To achieve an extended dynamic range in the projection data, we implemented a novel dynamic gain-switching mode. The read out charge amplifier feedback capacitance is changing dynamically in this mode, depending on the signal level. For this scan mode dedicated corrections in the offset and gain calibration are required. We compared image quality in terms of low contrast for both, the dynamic mode and the standard fixed gain mode. VCT scanners require different types of dose parameters. We measured the dose in a 16 cm CTDI phantom and free air in the scanners iso-center and defined a new metric for a VCT dose index (VCTDI). The dose for a high quality VCT scan of this prototype scanner varied between 15 and 40 mGy.

Our effort to implement a volumetric x-ray computed mammotomography (CmT) system dedicated to imaging breast disease comprises: demonstrated development of a quasi-monochromatic x-ray beam providing minimal dose and other optimal imaging figures of merit; new development of a compact, variable field-of-view, fully-3D acquisition gantry with a digital flat-panel detector facilitating more nearly complete sampling of frequency space and the physical breast volume; incorporation of iterative ordered-subsets transmission (OSTR) image reconstruction allowing modeling of the system matrix. Here, we describe the prototype 3D gantry and demonstrate initial system performance. Data collected on the prototype gantry demonstrate the feasibility of using OSTR with realistic reconstruction times. The gantry consists of a rotating W-anode x-ray tube using ultra-thick K-edge filtration, and an ~20x25cm² digital flat-panel detector located at <60cm SID. This source/detector combination can be shifted laterally changing the location of the central ray relative to the system center-of-rotation, hence changing the effective imaging field-of-view, and is mounted on a goniometric cradle allowing <50° polar tilt, then on a 360° azimuthal rotation stage. Combined, these stages provide for positioning flexibility in a banded region about a sphere, facilitating simple circle-plus-arc-like trajectories, as well as considerably more complex 3D trajectories. Complex orbits are necessary to avoid physical hindrances from the patient while acquiring the largest imaging volume of the breast. The system capabilities are demonstrated with fully-3D reconstructed images of geometric sampling and resolution phantoms, a fabricated breast phantom containing internal features of interest, and a cadaveric breast specimen. This compact prototype provides flexibility in dedicated, fully-3D CmT imaging of healthy and diseased breasts.

Cone-beam computed tomography (CBCT) presents a highly promising and challenging advanced application of flat-panel detectors (FPDs). The great advantage of this adaptable technology is in the potential for sub-mm 3D spatial resolution in combination with soft-tissue detectability. While the former is achieved naturally by CBCT systems incorporating modern FPD designs (e.g., 200 - 400 um pixel pitch), the latter presents a significant challenge due to limitations in FPD dynamic range, large field of view, and elevated levels of x-ray scatter in typical CBCT configurations. We are investigating a two-pronged strategy to maximizing soft-tissue detectability in CBCT: 1) front-end solutions, including novel beam modulation designs (viz., spatially varying compensators) that alleviate detector dynamic range requirements, reduce x-ray scatter, and better distribute imaging dose in a manner suited to soft-tissue visualization throughout the field of view; and 2) back-end solutions, including implementation of an advanced FPD design (Varian PaxScan 4030CB) that features dual-gain and dynamic gain switching that effectively extends detector dynamic range to 18 bits. These strategies are explored quantitatively on CBCT imaging platforms developed in our laboratory, including a dedicated CBCT bench and a mobile isocentric C-arm (Siemens PowerMobil). Pre-clinical evaluation of improved soft-tissue visibility was carried out in phantom and patient imaging with the C-arm device. Incorporation of these strategies begin to reveal the full potential of CBCT for soft-tissue visualization, an essential step in realizing broad utility of this adaptable technology for diagnostic and image-guided procedures.

Large area, real-time, amorphous selenium (a-Se) based Flat Panel Detectors (FPD) were recently equipped with low noise front end electronics. In full resolution, 14”x14” detectors (FPD14) and 9”x9” detectors (FPD9) show an electronic noise of 1400 electrons. To evaluate the positive impact of such low noise on image quality, a dedicated report on spatial characteristics (MTF, NPS and DQE) covering the low dose range from 0.6 μR to 12 μR per frame, will be presented in the first section of this paper. For one RQA5 beam quality, DQE corrected for lag extrapolated at zero spatial frequency was equal to 0.6 for quantum noise limited exposure and equal to 0.4 for 0.6 μR. Almost no difference was found between 1x1 and 2x2 resolution mode giving the opportunity to 1x1 fluoroscopy. Recent advances to reduce image temporal artifacts such as lag and ghost will make the second part of this paper. It is demonstrated that the most significant contribution to detector lag is coming from the PIN selenium structure. Above electric field of 10 V/μm charges release from traps following one x-ray exposure could not explain selenium lag. Active ghost correction based on deep trapped charge recombination was developed giving good preliminary results in showing no residual ghost for a high dose rate of 33 mR/min.

Amorphous selenium direct-conversion x-ray detectors have been used successfully for full field digital mammography (FFDM) and digital radiography (DR). Such detectors characteristically exhibit high spatial resolution and conversion efficiency that is a function of the applied electric field. At an electric field of 10 volts per micron, about 50 electron volts of photon energy are required to generate an electron-hole pair in a typical amorphous selenium x-ray conversion layer. At FFDM and DR imaging x-ray energies each absorbed photon can generate only about 250 to 1000 electron-hole pairs. Each absorbed x-ray photon is only contributing 4 x 10^-17 to 1.6 x 10^-16 coulombs of imaging charge. On the noise side, detectors operating at room temperature have a basic thermal (kTC) noise of 300 to 600 electrons per pixel from the image charge storage capacitor. Electronic noise from the front-end charge amplifier is also amplified by one plus the ratio of the TFT data line capacitance and the feedback capacitance of the charge amplifier. Medical imaging applications must therefore employ low noise thin film transistor (TFT) arrays, low data line capacitance and low noise charge integration amplifiers to achieve high signal-to-noise ratio (SNR) and detective quantum efficiency (DQE). To achieve quantum-noise limited imaging results with the lowest practical x-ray exposure dose, it is desirable to include an additional low-noise gain stage in the x-ray conversion layer. This is particularly important for the application of dynamic imaging or for tomosynthesis where x-ray dose per frame is very limited. A new structure for an amorphous selenium detector that employs an internal biased gain grid to cause avalanche-gain within the x-ray conversion layer is being proposed. A signal charge amplification of at least 10X can be achieved without introducing excessive noise. Quantum-limited image detection should then be attainable for even very low exposures.

The dependence of the x-ray sensitivity of a-Se based x-ray image detectors on repeated x-ray exposures is studied by considering deep trapping of charge carriers, trapped charges due to previous exposures, trap filling effects, recombination between trapped and drifting carriers, x-ray induced new deep trap center generation, space charge effects, and electric field dependent electron-hole pair creation energy. We simultaneously solve the continuity equations for both holes and electrons, trapping rate equations, and the Poisson’s equation across the photoconductor for a pulse x-ray exposure by the finite difference method. We also perform Monte Carlo Simulations of carrier transports and obtain almost identical results. The change in relative sensitivity (ghosting) as a function of cumulative x-ray exposures for different levels of trapping and different detector operating conditions are examined. The relative sensitivity decreases with increasing cumulated x-ray exposure. The amount of ghosting in a-Se detectors increases with decreasing applied electric field. The sensitivity reduction at negative bias is greater than at positive bias. The theoretical model shows a very good agreement with the experimental relative sensitivity vs. cumulative x-ray exposure characteristics. The comparison of the model with the experimental data reveals that the recombination between trapped and the oppositely charged drifting carriers and x-ray induced new deep trap centers are mainly responsible for the sensitivity reduction in biased a-Se-based x-ray detectors.

The purpose of this work was to evaluate the effect of detector element size on detection and discrimination of small objects in digital mammograms in the presence of quantum noise, stochastic breast structure {P (f) = K/f³)} and system (electronic) noise. Theoretical analysis was done using the Fourier domain approximation of Albert and Maidment (Medical Physics, Vol. 27, pp 2417-2434, 2000), including averaging over random but known signal locations and aliasing. Realistic CsI-based indirect detection system operating parameters were used with 4 detector element sizes (25, 50, 75 and 100 microns). Detailed results depended on selected input exposure and system noise level (10 mR and 100 photons/del equivalent photon input were used). Results will be described using size thresholds based on human decision performance estimates -- typical threshold SNRs were about 6 to 7 at 95% decision accuracy. The selected tasks and approximate threshold sizes were: searching for a microcalcification in a 1 cm² cluster (140 microns), two-alternative forced-choice microcalcification discrimination of shape (220 microns) and edge gradient (290 microns), and searching for a spiculation (cylinder model) around the perimeter of a known mass. MC task threshold variations with del size were small (usually less than 5% variation). Spiculation thresholds depended on length and surrounding tissue composition -- del size and system noise had little effect. For 15 mm spiculations, the threshold diameters were about 0.5 mm in fatty tissue and 1.3 mm in 50% glandular tissue.

We report on a set of tests that measure the performance of a-Si flat panel TFT arrays used in digital x-ray detectors. During production of high performance TFT panels for applications such as mammography it is important to verify the integrity and quality of the TFT array at progressive stages of production. Early identification of failing TFT arrays as well as continuous monitoring of the production process can result in early termination of poor quality panels, quick identification of the root cause of failures, and correction of process drift to prevent failures from occurring. We present results of a system designed to test the performance of a-Si TFT arrays during the production process. Metrics which are important to x-ray image quality were tested, including FET performance, pixel capacitance, storage capacitor lag and diode leakage. Functional tests were performed entirely on pixels in the imaging array using timing and biasing conditions that mimic x-ray illumination.

Scatter correction is an active research topic in cone beam computed tomography (CBCT) because CBCT (especially flat-panel detector (FPD) based) systems have large scatter-to-primary ratios. Scatter produces artifact and contrast reduction, and is difficult to model accurately. Direct measurement using a beam blocker array provides accurate scatter estimates. However, since the blocker array also blocks primary radiation, imaging requires a second (or subsequent) scan without the blocker array in place. This approach is inefficient in terms of scanning time and patient dose. To combine accurate scatter estimation and reconstruction into one single scan, a new approach based on an array of moving blockers has been developed. The blocker array moves from projection to projection, such that every detector pixel is not consecutively blocked during the data acquisition, and the missing primary data in the blocker shadows are estimated by interpolation. Using different blocker array trajectories, the algorithm has been evaluated through software phantom studies using Monte Carlo simulations and image processing techniques. Results show that this approach is able to greatly reduce the effect of scatter in the reconstruction. By properly choosing blocker distance and primary data interpolation method, the mean square error of the reconstructed image decreases from 32.3% to 1.13%, and the induced visual artifacts are significantly reduced when a raster-scanning blocker array trajectory is used. Further analysis also shows that artifact arises mostly due to inaccurate scatter estimates, rather than due to interpolation of the primary data.

This study deals with a systematic assessment of the potential of different schemes for computerized scatter correction in flat detector based cone-beam X-ray computed tomography. The analysis is based on simulated scatter of a CT image of a human head. Using a Monte-Carlo cone-beam CT simulator, the spatial distribution of scattered radiation produced by this object has been calculated with high accuracy for the different projected views of a circular tomographic scan. Using this data and, as a reference, a scatter-free forward projection of the phantom, the potential of different schemes for scatter correction has been evaluated. In particular, the ideally achievable degree of accuracy of schemes based on estimating a constant scatter level in each projection was compared to approaches aiming at estimation of a more complex spatial shape of the scatter distribution. For each scheme, remaining cupping artifacts in the reconstructed volumetric image were quantified and analyzed. It was found that already accurate estimation of a constant scatter level for each projection allows for comparatively accurate compensation of scatter-caused artifacts.

Scattered radiation is a major source of image degradation and nonlinearity in flat detector based cone-beam CT. Due to the bigger irradiated volume the amount of scattered radiation in true cone-beam geometry is considerably higher than for fan beam CT. This on the one hand reduces the signal to noise ratio, since the additional scattered photons contribute only to the noise and not to the measured signal, and on the other hand cupping and streak artifacts arise in the reconstructed volume. Anti-scatter grids composed of lead lamellae and interspacing material decrease the SNR for flat detector based CB-CT geometry, because the beneficial scatter attenuating effect is overcompensated by the absorption of primary radiation. Additionally, due to the high amount of scatter that still remains behind the grid, cupping and streak artifacts cannot be reduced sufficiently. Computerized scatter correction schemes are therefore essential for achieving artifact-free reconstructed images in cone-beam CT. In this work, a fast model based scatter correction algorithm is proposed, aiming at accurately estimating the level and spatial distribution of scattered radiation background in each projection. This will allow for effectively reducing streak and cupping artifacts due to scattering in cone-beam CT applications.

The expanding set of CT clinical applications demands increased attention to obtaining the maximum image quality at the lowest possible dose. Pre-patient beam shaping filters provide an effective means to improve dose utilization. In this paper we develop and apply characterization methods that lead to a set of filters appropriately matched to the patient. We developed computer models to estimate image noise and a patient size adjusted CTDI dose. The noise model is based on polychromatic X-ray calculations. The dose model is empirically derived by fitting CTDI style dose measurements for a demographically representative set of phantom sizes and shapes with various beam shaping filters. The models were validated and used to determine the optimum IQ vs dose for a range of patient sizes. The models clearly show that an optimum beam shaping filter exists as a function of object diameter. Based on noise and dose alone, overall dose efficiency advantages of 50% were obtained by matching the filter shape to the size of the object. A set of patient matching filters are used in the GE LightSpeed VCT and Pro³² to provide a practical solution for optimum image quality at the lowest possible dose over the range of patient sizes and clinical applications. Moreover, these filters mark the beginning of personalized medicine where CT scanner image quality and radiation dose utilization is truly individualized and optimized to the patient being scanned.

Properties and performance of digital X-ray detectors for medical imaging can be studied by Monte Carlo simulations. Most simulations of such detectors simplify the setup by only taking the conversion layer into account neglecting everything behind. For hybrid detectors with Si as the conversion layer, such as the Medipix2 chip less photons are absorbed at higher photon energies in the conversion layer and thus may reach the detector ASIC including its bump bonds. For photon energies above the K-edges of the backscatter materials, fluorescence may occur. The fluorescence photons can have relatively long ranges and thus have a great impact on the MTF of the detector decreasing its spatial resolution. They also add noise to the detector decreasing the overall signal-difference-to-noise-ratio (SDNR). In our study we simulated the line spread functions (LSF) for photon-counting pixel detectors by Monte Carlo simulations, implementing the detectors in detail. We used the program ROSI (ROentgen SImulation) which is based on the well-established EGS4 algorithm. The appropriate MTFs were calculated by FFT. We show that internal backscattering, especially from Sn bump bonds, contributes to the so-called low-frequency drop of the MTF. For a 300 μm Si absorber on the Medipix2 chip, backscattering contributes up to 10% to the detected signal. This strongly decreases contrast by adding additional noise. Therefore, we also investigated the amount of noise added by internal backscattering.

We are investigating the advantage of scatter removal by the slot-scanning method compared to antiscatter grids. We carry out model calculations for the signal-to-noise ratio simulating different geometrical settings for slot-scan systems. The results are compared with those for standard nonscanning mammography systems with and without anti-scatter grid. Monte Carlo simulations are performed in order to get a realistic amount of scatter radiation as input for the model estimates. We present the results as function of the compressed breast thickness equivalent to the scatter fraction. It is demonstrated that a perfect slot-scan system with 100% transmission of primary radiation and 100% suppression of scattered radiation improves SNR², and correspondingly reduces dose, by a factor of less than 1.8, compared with conventional anti-scatter grids and otherwise the same detector DQE. For realistic geometry the advantage is considerably smaller. The advantage of scatter removal by employing a slot-scanning method is moderate because the scatter fraction is relatively low in mammography. For breast thickness up to 5 cm it turns out that it is advantageous to work without a grid due to the low scatter fraction, which questions a scatter reduction method in that region at all. The model can be used as a simple design tool.

Flat panel detector-based cone beam CT (FPD-CBCT) imaging system prototypes have been constructed based on modified clinical CT scanners (a modified GE 8800 CT system and a modified GE HighSpeed Advantage (HSA) spiral CT system) each with a Varian PaxScan 2520 imager. The functions of the electromechanical and radiographic subsystems of the CT system were controlled through specially made hardware, software and data acquisition modules to perform animal cone beam CT studies. Small animal (mouse) imaging studies were performed to demonstrate the feasibility of an optimized CBCT imaging system to have the capability to perform longitudinal studies to monitor the progression of cancerous tumors or the efficacy of treatments. Radiographic parameters were optimized for fast (~10 second) scans of live mice to produce good reconstructed image quality with dose levels low enough to avoid any detectable radiation treatment to the animals. Specifically, organs in the pelvic region were clearly imaged and contrast studies showed the feasibility to visualize small vasculature and space-filling bladder tumors. In addition, prostate and mammary tumors were monitored in volume growth studies.

While mammography is the gold standard for breast cancer screening worldwide, it is widely recognized that mammography has limitations, especially in women with dense breasts. In response to the need for a more sensitive approach to breast cancer screening, a CT scanner specifically for breast imaging in the pendant geometry was designed, fabricated, and is currently in clinical evaluation. The spatial resolution and noise properties are discussed, and breast images from a normal volunteer and a patient with breast cancer demonstrate very promising breast CT image quality from a qualitative perspective.

With the development of new technologies, computed tomography (CT) is becoming a strong candidate for non-invasive imaging based tool for cardiac disease assessment. One of the challenges of cardiac CT is that a typical scan involves a breath hold period consisting of several heartbeats, about 20 sec with scanners having a longitudinal coverage of 2 cm, and causing the image quality (IQ) to be negatively impacted since beat to beat variation is high likely to occur without any medication, e.g. beta blockers. Because of this and the preference for shorter breath hold durations, a CT scanner with a wide coverage without the compromise in the spatial and temporal resolution of great clinical value. In this study, we aimed at determining the optimum scan duration and the delay relative to beginning of breath hold, to achieve high IQ. We acquired EKG data from 91 consecutive patients (77 M, 14 F; Age: 57 ± 14) undergoing cardiac CT exams with contrast, performed on LightSpeed 16 and LightSpeed Pro¹⁶. As an IQ metric, we adopted the standard deviation of "beat-to-beat variation" (stdBBV) within a virtual scan period. Two radiologists evaluated images by assigning a score of 1 (worst) to 4 best). We validated stdBBV with the radiologist scores, which resulted in a population distribution of 9.5, 9.5, 31, and 50% for the score groups 1, 2, 3, and 4, respectively. Based on the scores, we defined a threshold for stdBBV and identified an optimum combination of virtual scan period and a delay. With the assumption that the relationship between the stdBBV and diagnosable scan IQ holds, our analysis suggested that the success rate can be improved to 100% with scan durations equal or less than 5 sec with a delay of 1 - 2 sec. We confirmed the suggested conclusion with LightSpeed VCT (GE Healthcare Technologies, Waukesha, WI), which has a wide longitudinal coverage, fine isotropic spatial resolution, and high temporal resolution, e.g. 40 mm coverage per rotation of 0.35 sec. Under the light of this study, LightSpeed VCT lends itself to be a clinically tested unique platform to achieve routine cardiac imaging.

The basic VRX technique boosts spatial resolution of a CT scanner in the scan plane by two or more orders of magnitude by reducing the angle of incidence of the x-ray beam with respect to the detector surface. A four-arm Variable-Resolution X-ray (VRX) detector has been developed for CT scanning. The detector allows for "target imaging" in which an area of interest is scanned at higher resolution than the remainder of the subject, yielding even higher resolution for the focal area than that obtained from the basic VRX technique. The new VRX-CT detector comprises four quasi-identical arms each containing six 24-cell modules (576 cells total). The modules are made of individual custom CdWO4 scintillators optically-coupled to custom photodiode arrays. The maximum scan field is 40 cm for a magnification of 1.4. A significant advantage of the four-arm geometry is that it can transform quickly to the two-arm, or even the single-arm geometry, for comparison studies. These simpler geometries have already been shown experimentally to yield in-plane CT detector resolution exceeding 60 cy/mm (<8μ) for small fields of view. Geometrical size and resolution limits of the target VRX field are calculated. Two-arm VRX-CT data are used to simulate and establish the feasibility of VRX CT target imaging. A prototype target VRX-CT scanner has been built and is undergoing initial testing.

Tracking and targeting the tumors are simultaneous processes in the image-guided radiotherapy (IGRT); this is expected to boost the efficiency, the reliability, and the safety in the treatment. Varian Medical Systems (VMS) has already produced and installed the first IGRT machine; the device comprises the VMS Clinac equipped with the On-Board Imager (OBI) component. Cone-beam CT (CBCT) imaging, one of the options of the OBI machine, aims at high-quality volumetric reconstruction. A number of calibrations are needed in order to operate our CT-imaging machines properly; they ensure that the machine components are properly aligned, the mechanical distortions are small, and yield important output that is used in the reconstruction of the actual scan data. The geometrical calibration is achieved by using a needle phantom. In order to increase the dynamic range of our imager (hence, to obtain reliable information simultaneously in the high- and the low-attenuation areas of the irradiated object), VMS has developed a dual-gain mode. Next on our agenda is the suppression of (ring, streak, and beam-hardening) artefacts in our reconstructed images and the further development of our detectors in order to remove patterns relating to lag and ghosting effects.

An indirect flat-panel imager (FPI) with avalanche gain is being investigated for low-dose x-ray imaging. It is made by optically coupling a structured x-ray scintillator CsI(Tl) to an amorphous selenium (a-Se) avalanche photoconductor called HARP. The final electronic image can be read out using either an array of thin film transistors (TFT) or field emitters (FE). The advantage of the proposed detector is its programmable gain, which can be turned on during low dose fluoroscopy to overcome electronic noise, and turned off during high dose radiography to avoid pixel saturation. This paper investigates the important design considerations for HARP such as avalanche gain, which depends on both the thickness d_Se and the applied electric field E_Se. To determine the optimal design parameter and operational conditions for HARP, we measured the E_Se dependence of both avalanche gain and optical quantum efficiency of an 8 μm HARP layer. The results were applied to a physical model of HARP as well as a linear cascaded model of the FPI to determine the following x-ray imaging properties in both the avalanche and non-avalanche modes as a function of E_Se: (1) total gain (which is the product of avalanche gain and optical quantum efficiency); (2) linearity; (3) dynamic range; and (4) gain non-uniformity resulting from thickness non-uniformity. Our results showed that a HARP layer thickness of 8 μm can provide adequate avalanche gain and sufficient dynamic range for x-ray imaging applications to permit quantum limited operation over the range of exposures needed for radiography and fluoroscopy.

Simulations of digital imaging systems based on scintillator screens usually employ a Poisson model for the phosphor conversion gain. However, the statistics of the scintillation output are determined by complex phenomena that involve many sources of variability including inhomogeneities in the crystalline and screen structure, variations in the deposited energy for each primary quantum available for excitation, variations in the relationship between radiate and non-radiative decay processes, energy dependencies in the conversion gain variance, and spread of secondary quanta. We use a combined x-ray/electron/optical Monte Carlo code to study the statistics of the scintillation output in columnar phosphors. The simulation code is the result of merging the x-ray transport code PENELOPE and the optical transport code DETECT-II. Using an improved geometric model for the columnar structure, we present results concerning pulse-height spectra of the scintillation output (and corresponding Swank factors) as a function of x-ray energy. This study improves our understanding of the underlying causes of conversion gain variations and should facilitate more accurate simulation efforts for the investigation and optimization of image acquisition systems based on scintillator screens.

Image statistics are usually modeled as Poisson in γ-ray imaging and as Gaussian in x-ray imaging. In nuclear medicine, event-driven detectors analyze the pulses from every absorbed gamma photon individually; the resulting images rigorously obey Poisson statistics but are approximately Gaussian when the mean number of counts per pixel is large. With integrating detectors, as in digital radiography, each x-ray photon makes a contribution to the image proportional to its pulse height. One pixel senses many photons in long exposures, so the image statistics approach Gaussian by the central limit theorem (CLT). If the exposure time is short enough, however, each pixel will usually respond to no more than one photon, and we can separate individual photons for position estimation. Integrating detectors are therefore event-driven when we use many short-exposure frames rather than one long exposure. In intermediate exposures, the number of photons in one pixel is too small to invoke CLT and apply Gaussian statistics, yet too large to identify individual photons and apply Poisson statistics. In this paper, we analyze the image quality in this intermediate case. Image quality is defined for detection tasks performed by the ideal observer. Because the frames in a data set are independent of each other, the probability density function (PDF) of the whole data set is a product of the frame PDFs. The log-likelihood ratio λ of the ideal observer is thus a sum across the frames and has Gaussian statistics even with non-Gaussian images. We compare the ideal observer's performance with the Hotelling observer's performance under this approximation. A data-thresholding technique to improve Hotelling observer's performance is also discussed.

Image artifacts caused by a temporally delayed response of a back illuminated photodiode deployed in a CT detector were studied. The temporal response pattern, characterized by a finite rise and fall time, is an intrinsic property of a photodiode. Generally, electron-hole pairs generated in the diode take time to diffuse to contacts where they get finally registered. In the case of backilluminated diodes, diffusion time is significantly prolonged, since photons hit the diode on the back. Electrons or holes only contribute to the signal, if they travel the full distance to the frontend, where contacts are located. To study the temporal behavior of a back illuminated photodiode a computer model for a standard third generation CT scanner was devised and simulations were carried out. Resulting image artifacts were quantified for various phantom and photodiode parameters. Simulations and theory demonstrate that for a given phantom, artifacts scale with rise/fall time and the angular speed of the scanner.

This paper explores the potential of flat panel detectors in sub-second CT scanning applications. Using a PaxScan 4030CB with 600um thick CsI(Tl), a central section of the panel (16 to 32 rows), was scanned at frame rates up to 1000fps. Using this platform, fundamental issues related to high speed scanning were characterized. The offset drift of the imager over 60 seconds was found to be less than 0.014 ppm/sec relative to full scale. The gain stability over a 10 hour period is better than +/- .45%, which is at the resolution limit of the measurement. Two different types of lag measurements were performed in order to separate the photodiode array lag from the CsI afterglow. The panel lag was found to be 0.41% 1^st frame and 0.054% 25^th frame at 1000fps. The CsI(Tl) afterglow, however, is roughly an order of magnitude higher, dominating the lag for sub-second scans. At 1000fps the 1^st frame lag due to afterglow was 3.3% and the 25^th frame lag was 0.34%. Both the lag and afterglow are independent of signal level and each follows a simple power law evolution versus time. Reconstructions of anatomical phantoms and the CATPHAN 500 phantom are presented. With a 2 second, 1200 projection scan of the CATPHAN phantom at 600fps in 32 slice mode, using 120kVp and CTDI₁₀₀ of 43.2mGy, 0.3% contrast resolution for a 6mm diameter target, can be visualized. In addition, 15lp/cm spatial resolution was achieved with a 2mm slice and a central CTDI₁₀₀ of 10.8mGy.

Cassette based computed radiography systems have continued to evolve in parallel with integrated, instant readout digital radiography DR systems. The image quality of present day computed radiography systems is approaching its theoretical limits but is still significantly inferior to DR. Our overall aim is to identify the fundamental limitations in computed radiography performance. This will provide a basis for the development of new approaches to improve photostimulable phosphor based computed radiography systems based either on cassettes or integrated systems. In this fundamental work flare has been evaluated by use of the disk transfer function. It is found that there is a significant, previously unnoticed low frequency drop in the modulation transfer function. However, it is not yet fully resolved if the measured flare is due solely to scattering of light in the readout system or whether scattered x-rays in the imaging plate cassette enhance it. Further work is necessary to resolve this issue. The likely effect of flare on proposed new computed radiography methods is also explored.

A generalized, objective image quality measure can be defined for X-ray based medical projection imaging: the spatial frequency-dependent signal-to-noise ratio SNR = SNR(u,v). This function includes the three main image quality parameters, i.e. spatial resolution, object contrast, and noise. The quantity is intimately related to the DQE concept, however its focus is not to characterize the detector, but rather the detectability of a certain object embedded into a defined background. So also effects from focus size and radiation scatter can be quantified by this method. The SNR(u,v) is independent of basic linear post-processing steps such as appropriate windowing or spatial filtering. The consideration of the human visual system is beyond the scope of this concept. By means of this quantity, different X-ray systems and setups can be compared with each other and with theoretical calculations. Moreover, X-ray systems (i.e. detector, beam quality, geometry, anti-scatter grid, basic linear post-processing steps etc.) can be optimized to deliver the best object detectability for a given patient dose. In this paper SNR(u,v) is defined using analytical formulas. Furthermore, we demonstrate how it can be applied with a test phantom to a typical flat panel detector system by a combination of analytical calculations and Monte Carlo simulations. Finally the way this function can be used to optimize an X-ray imaging device is demonstrated.

Standard objective parameters such as MTF, NPS, NEQ and DQE do not reflect complete system performance, because they do not account for geometric unsharpness due to finite focal spot size and scatter due to the patient. The inclusion of these factors led to the generalization of the objective quantities, termed GMTF, GNNPS, GNEQ and GDQE defined at the object plane. In this study, a commercial x-ray image intensifier (II) is evaluated under this generalized approach and compared with a high-resolution, ROI microangiographic system previously developed and evaluated by our group. The study was performed using clinically relevant spectra and simulated conditions for neurovascular angiography specific for each system. A head-equivalent phantom was used, and images were acquired from 60 to 100 kVp. A source to image distance of 100 cm (75 cm for the microangiographic system) and a focal spot of 0.6 mm were used. Effects of varying the irradiation field-size, the air-gaps, and the magnifications (1.1 to 1.3) were compared. A detailed comparison of all of the generalized parameters is presented for the two systems. The detector MTF for the microangiographic system is in general better than that for the II system. For the total x-ray imaging system, the GMTF and GDQE for the II are better at low spatial frequencies, whereas the microangiographic system performs substantially better at higher spatial frequencies. This generalized approach can be used to more realistically evaluate and compare total system performance leading to improved system designs tailored to the imaging task.

The detective quantum efficiency (DQE) is regarded as a suitable parameter to assess the global imaging performance of an x-ray detector. However, residual signals increase the signal-to-noise ratio and therefore artificially increase the measured DQE compared to a lag-free system. In this paper, the impact of lag on the DQE is described for two different sources of lag using linear system models. In addition to the commonly used temporal filtering model for trapping, an increase of the dark current is considered as another potential source of lag. It is shown that the assumed lag model has a crucial impact on the choice of an adequate lag estimation method. Examples are given using the direct conversion material PbO. It turns out that the most general approach is the evaluation of the temporal noise power spectrum. A new algorithm is proposed for the crucial issue of robustly estimating the power spectrum at frequency zero.

A variety of incompatible speed metrics for digitally acquired projection images are currently provided by system vendors. As digital acquisition has become more widely adopted, the need for a universal vendor-independent speed metric is increasingly evident. This paper proposes a method for defining the speed of digital projection images that can be readily implemented on any system that acquires a projection x-ray image and produces a display-ready image. Radiographic speed for screen-film combinations is currently defined by ISO in terms of the exposure needed to produce a net density of one on the developed film. An analogous speed method is proposed for digital images. It requires that the system produce an original image (calibrated in terms of the relationship between system response and exposure for the standard set of x-ray beam qualities defined by ISO-9236-1) and a display-ready image. The speed is computed in terms of the median pixel value in the original image that corresponds to a reference pixel value in the display-ready image. The exposure response of selected digital radiography acquisition systems has been measured for the ISO beam qualities. The proposed speed metric was computed for a representative suite of digital images and correlated well with a currently available vendor-specific speed metric. Changes in patient exposure and image-processing parameters affect the speed metric in the appropriate way. In conclusion, the proposed speed metric provides a vendor-independent definition of speed for digitally acquired projection radiographs that is consistent with current speed standards for screen-film radiography and applicable to all currently available digital acquisition systems.

A digital radiography system comprised of a large field of view (43x43cm) high luminance CsI scintillator, optically coupled to a 4096x4096 element CCD sensor with 12:1 demagnification was evaluated by measuring the modulation transfer function (MTF), noise power spectrum (NPS), and detective quantum efficiency (DQE). Detector evaluation was performed using IEC standard #62220-1 methodologies for beam quality. In this study, RQA-5 (21 mm Al added filtration, 74 kVp, 7.1 mm half value layer (HVL)) and RQA-9 (40 mm Al added filtration, 119 kVp, 11.5 mm HVL) qualities were used at several incident exposures from <0.1 mR to >50 mR. Two detector modes of operation included high resolution (HR) and high efficiency (HE), with 108 and 216 μm pixel dimensions, respectively. The detector system responded to 60 mR incident exposure prior to saturation for the HR mode and up to 30 mR in the HE mode. The pre-sampled MTF(f) had 50% modulation at 0.95 mm^-1 (HR) and 0.85 mm^-1 (HE); and 10%MTF(f) was reached at 2.4 mm^-1 (HR) and 2.0 mm^-1 (HE). At a frequency of 0.5 mm^-1, the DQE was 40% to 50%, and at 1 mm^-1 was 12% to 20% for HR and HE modes, respectively. The DQE at low exposures was substantially better for the HE mode. Little dependence of the DQE on beam energy was found, but the RQA 9 beam had lower values. Above a frequency of 2 mm^-1 the DQE dropped to zero, attributed to low MTF. Results suggest that patient radiation exposures equivalent or better than a conventional 400 speed screen-film detector can be achieved for many imaging procedures with sufficient SNR and spatial resolution required for a wide range of diagnostic radiography applications.

In mammography, there is an optimal photon energy and current time product that produce the required image quality at the minimal dose. The task of an automatic exposure control (AEC), in full field digital mammography (FFDM) is to minimize the dose by using optimized exposure settings. Each point in a mammogram has different radiological thickness. A conventional AEC samples the thickness in some regions to set the current time product and possibly also the beam quality. We define an ideal AEC as one that optimizes the beam quality and exposure in each point to produce a constant contrast-to-noise ratio (CNR) of structures of interest throughout the image. This paper presents the results from a theoretical evaluation of an AEC proposed for a scanning photon-counting FFDM system. The geometry enables the AEC to use information from the leading detector line to adjust the scan velocity during the scan. Thus, the irradiation can be better optimized in the scanning-direction as compared to a conventional AEC. The scan time is further reduced by increased velocity over sections that contain no tissue. The results are quantified in terms of reduction of entrance dose and scan time. The presented AEC is compared to an ideal AEC, a conventional AEC and is also benchmarked against an ideal regulator. The effect of the detector width is evaluated. Compared to a conventional AEC, both evaluated on a set of 266 mammograms, the ideal AEC would reduce the entrance dose by 39% on average while the proposed AEC for scanning systems reduces the entrance dose by 10-20% and scan-time by 25-32% on average, depending on detector width.

This study evaluated the physical performance of a selenium-based direct full-field digital mammography prototype detector (Siemens Mammomat Novation^DR), including the pixel value vs. exposure linearity, the modulation transfer function (MTF), the normalized noise power spectrum (NNPS), and the detective quantum efficiency (DQE). The current detector is the same model which received an approvable letter from FDA for release to the US market. The results of the current prototype are compared to those of an earlier prototype. Two IEC standard beam qualities (RQA-M2: Mo/Mo, 28 kVp, 2 mm Al; RQA-M4: Mo/Mo, 35 kVp, 2 mm Al) and two additional beam qualities (MW2: W/Rh, 28 kVp, 2 mm Al; MW4: W/Rh, 35 kVp, 2 mm Al) were investigated. To calculate the modulation transfer function (MTF), a 0.1 mm Pt-Ir edge was imaged at each beam quality. Detector pixel values responded linearly against exposure values (R² 0.999). As before, above 6 cycles/mm Mo/Mo MTF was slightly higher along the chest-nipple axis compared to the left-right axis. MTF was comparable to the previously reported prototype, with slightly reduced resolution. The DQE peaks ranged from 0.71 for 3.31 μC/kg (12.83 mR) to 0.4 for 0.48 μC/kg (1.86 mR) at 1.75 cycles/mm for Mo/Mo at 28 kVp. The DQE range for W/Rh at 28 kVP was 0.81 at 2.03 μC/kg (7.87 mR) to 0.50 at 0.50 μC/kg (1.94 mR) at 1 cycle/mm. NNPS tended to increase with greater exposures, while all exposures had a significant low-frequency component. Bloom and detector edge artifacts observed previously were no longer present in this prototype. The new detector shows marked noise improvement, with slightly reduced resolution. There remain artifacts due to imperfect gain calibration, but at a reduced magnitude compared to a prototype detector.

The growing importance of the “European Protocol for the Quality Control of the Physical and Technical Aspects of Mammography Screening, Part B: Digital Mammography” dictates the need to understand the prescribed threshold contrast sensitivity test. Observers following a 4-AFC paradigm, report the location of disks varying in contrast and diameter on multiple images of a CDMAM or similar phantom. Analysis provides a contrast threshold for each disk diameter. The goals of this study were to quantify the performance of new observers, compare it to published results, compare visual scoring with software scoring of the same images, and to quantify the major sources of variability. Digital phantom images, visual scorings by four expert readers, and CDCOM software were downloaded from the EUREF website. These images were read on a 3M Barco flat-panel monitor by 13 observers and scored by CDCOM. Scores were analyzed using the published method from the CDMAM-phantom 3.4 manual and a signal detection theory-based method. The average contrast sensitivities of the 13 study observers generally exceeded the published values by ~10%. The 95% confidence limits for the mean of 6 images from the published data vary from ±20.2% to ±41.8% of their respective means, the average being 31.2%. The average confidence limit for selected study observers is ±36%. Comparisons between software and human observer results using the prescribed method of analysis-revealed marked differences, particularly for small diameter targets. These differences are mitigated by signal-detection-theory analysis of both datasets. The large inter-observer variability and the substantial time required for human scoring supports the need to qualify a readily available software solution.

The image quality of needle-image-plate (NIP) Computed Radiography (CR) scanners based on ScanHead technology was optimized. In order to get the best image quality for different applications, the influence of the phosphor layer thickness on the detective quantum efficiency (DQE) for different beam qualities was investigated. We compared a cassette-based, reflective CR-NIP-scanner to a new, transmissive flat-panel CR scanner with fixed, mounted NIP. The image quality was analyzed by DQE- and modulation transfer function (MTF) measurements supported by an observer study. The NIP systems reached DQE values up to three times higher than that of high-quality, state of the art CR scanners independent of the scanning principle. This allows a dose reduction by a factor of two to three without loss of image quality for both scanning systems. For high tube voltages, the variation of the phosphor layer thickness results in a DQE maximum at relatively large thicknesses. For lower tube voltages the DQE is less dependent on the layer thickness, reaching excellent values already at considerably lower thicknesses. Consequently, CR scanners can be adapted to different applications by using NIPs with different thicknesses. This could be easily realized for the cassette based system, but not for the flat-panel system with fixed IP. The latter demands a compromise with respect to the phosphor thickness, to yield superior image quality for all applications.

The digital mammography system evaluated here is a digital full-field phase contrast mammography (PCM) system that employs a practical molybdenum x-ray tube with a 0.1mm focal spot size. By using custom apparatus to position 14"x17" photostimulable phosphor plate 49cm from the object, a distance of 65cm was obtained between the object and the focal spot of the x-ray tube. A plate optimized for the PCM system acquired images magnified 1.75 times with a 14"x17" and was scanned at a sampling pitch of 43.75μm by using a CR system. The diagnostic images were reduced in printing to actual object size with a pixel size of 25μm on 8"x10" dry-processed film with a maximum density of 4. This study compares the performance of the system with that of a usual contact CR mammography system.

Dual-energy (DE) imaging is a promising x-ray modality for the screening and early detection of lung cancer but has seen limited application primarily due to the lack of an adequate image detector. Recent development of flat-panel detectors (FPDs) for advanced imaging applications provide a promising technology for DE imaging, and a theoretical framework to quantify the imaging performance of FPD-based DE imaging systems is useful for system design and optimization. Traditional methods employed to describe imaging performance in radiographic systems [i.e., detective quantum efficiency (DQE) and noise-equivalent quanta (NEQ)] are extended in this paper to DE imaging systems using FPDs. To quantify the essential advantage imparted by DE imaging, we incorporate a spatial-frequency-dependent “anatomical noise” term associated with overlying structures to yield the generalized DQE and NEQ. We estimate anatomical noise in DE images through measurements using an anthropomorphic chest phantom and parameterize the measurements using a 1/f model. Cascaded systems analysis of the generalized NEQ is shown to reveal the tradeoffs between anatomical noise and quantum noise in DE image reconstructions. The generalized dual-energy NEQ is combined with idealized task functions to compute the detectability index, providing an estimate of ideal observer performance in a variety of detection and discrimination tasks. The generalized analysis is employed to investigate optimal tissue cancellation and kVp selection as a function of dose and imaging task.

A prototype breast tomosynthesis system has been developed, allowing a total angular view of ±25°. The detector used in this system is an amorphous selenium direct-conversion digital flat-panel detector suitable for digital tomosynthesis. The system is equipped with various readout sequences to allow the investigation of different tomosynthetic data acquisition modes. In this paper, we will present basic physical properties -- such as MTF, NPS, and DQE -- measured for the full resolution mode and a binned readout mode of the detector. From the measured projections, slices are reconstructed employing a special version of filtered backprojection algorithm. In a phantom study, we compare binned and full resolution acquisition modes with respect to image quality. Under the condition of same dose, we investigate the impact of the number of views on artifacts. Finally, we show tomosynthesis images reconstructed from first clinical data.

Digital tomosynthesis mammography algorithms allow reconstructions of arbitrary planes in the breast from limited-angle series of projection images as the x-ray source moves along an arc above the breast. Though several tomosynthesis algorithms have been proposed, no complete comparison of the methods has previously been conducted. This paper presents an analysis of impulse response for four different tomosynthesis mammography reconstruction algorithms. Simulated impulses at different 3-D locations were simulated to investigate the sharpness of reconstructed in-plane structures and to see how effective each algorithm is at removing out-of-plane blur. Datasets with 41, 21 and 11 projection images of the impulse were generated with a total angular movement of +/- 10 degrees of the simulated x-ray point source. Four algorithms, including shift-and-add method, Niklason algorithm, filtered back projection (FBP), and matrix inversion tomosynthesis (MITS) are investigated. Compared with shift-and-add algorithm and Niklason method, MITS and FBP performed better for in-plane response and out-of-plane blur removal. MITS showed better out-of-plane blur removal in general. MITS and FBP performed better when projection numbers increase.

We have developed a breast tomosynthesis system utilizing a selenium-based direct conversion flat panel detector. This prototype system is a modification of Selenia, Hologic’s full field digital mammography system, using an add-on breast holding device to allow 3D tomosynthetic imaging. During a tomosynthesis scan, the breast is held stationary while the x-ray source and detector mounted on a c-arm rotate continuously around the breast over an angular range up to 30 degrees. The x-ray tube is pulsed to acquire 11 projections at desired c-arm angles. Images are reconstructed in planes parallel to the breastplate using a filtered backprojection algorithm. Processing time is typically 1 minute for a 50 mm thick breast at 0.1 mm in-plane pixel size, 1 mm slice-to-slice separation. Clinical studies are in progress. Performance evaluations were carried out at the system and the subsystem levels including spatial resolution, signal-to-noise ratio, spectra optimization, imaging technique, and phantom and patient studies. Experimental results show that we have successfully built a tomosynthesis system with images showing less structure noise and revealing 3D information compared with the conventional mammogram. We introduce, for the first time, the definition of “Depth of Field” for tomosynthesis based on a spatial resolution study. This parameter is used together with Modulation Transfer Function (MTF) to evaluate 3D resolution of a tomosynthesis system as a function of system design, imaging technique, and reconstruction algorithm. Findings from the on-going clinical studies will help the design of the next generation tomosynthesis system offering improved performance.

Our goal is to evaluate diagnostic digital breast tomosynthesis and ultrasound imaging clinical value in detecting and diagnosing early stage breast cancers. Determine if fusion imaging would decrease the number of biopsies and reduce further patient workup otherwise required to establish a definitive diagnosis. This paper presents the clinical results based on the study conducted at Helsinki University Central Hospital. Presentation demonstrates clinical dual modality images and results. Tomosynthesis of amorphous selenium based full field digital mammography system will be also presented. Forty asymptomatic women enrolled in the study based on prior identification of suspicious findings on screening mammograms where the possibility of breast cancer could not be excluded. Abnormal screening mammogram findings included tumor-like densities, parenchymal asymmetries and architectural distortions. Eight women were operated and 32 were not referred for surgery. Those cases, which were operated, three lesions represented ductal carcinoma in situ, two ductal carcinomas, one atypical ductal hyperplasia, one fibroadenoma and one radial scar. The 32 not operated cases revealed to be benign or superimposition of normal parenchymal breast tissue. The cases were returned to biennial screening. Ultrasound did not show clearly any lesions, but using tomosynthesis and ultrasound together we were able to analyze and locate the lesions exactly. Special tomosynthesis improves overall lesion detection and analysis. The value of tomosynthesis and ultrasound fusion imaging will be to provide additional clinical information in order to improve decision making accuracy to either confirm or exclude a suspected abnormality and in particular detect small breast cancers.

Digital breast tomosynthesis promises solutions to many of the problems currently associated with projection mammography, including elimination of artifactual densities due to the superposition of normal tissues and increasing the conspicuity of true lesions that would otherwise be masked by superimposed normal tissue. We have investigated tomosynthesis using 45 photon counting, orientation sensitive, linear detectors which are precisely aligned with the focal spot of the x ray source. The x-ray source and the digital detectors are scanned in a continuous motion across the object (patient); each linear detector collecting an image at a distinct angle. Simulations of the imaging system were performed to evaluate the effect of: (1) the range of angles over which projection images are acquired; and (2) the number of projection images acquired used in the tomosynthetic reconstruction. Two different simulations were evaluated; the first was a numerical simulation of a tungsten wire; the second consisted of tomosynthetic reconstructions of a cadaveric rabbit, in which the number and/or range of projection angles was varied. We have shown, analytically and through these simulations, that both the use of more projection angles and the use of a larger range of projection angles improve the image quality of tomosynthetic image reconstructions. The use of a photon-counting x-ray detector system allows us to consider image acquisition geometries with a large number of projection angles, as there is no additive detector noise to degrade the projection or reconstructed images. The maximum number of projection angles and the range of projections angles do have upper practical limits; the range of projection angles is determined predominantly by the detector element size.

Cone-beam CT reconstruction can be performed at lower integral dose, by using a non-uniform beam filter between the x-ray source and the patient to obtain good image quality within an ROI with minimal artifacts. To evaluate the method, a human head phantom was placed on a rotary stage. Cone-beam projection images of the phantom were obtained with and without an ROI filter (dose reduction factor ~7). A mapping function was established to equalize the intensity outside the ROI (to compensate for the attenuation by the filter) to the intensity inside by assuming that those features lying both inside and outside very close to the edge of the ROI are the same. Reconstructed images were obtained using equalized projection images for 2 cases: one in which the outside region was smoothed using an averaging filter and the other with no smoothing outside. In addition, a third case was simulated by calculating the average pixel value inside the ROI for each image and assigning this value to all pixels outside the ROI for that image. The images were then back projected using a Feldkamp algorithm. We found that the three cases yield results inside the ROI comparable to those obtained using FFOV projections. In addition, the ROI filter reconstruction with smoothing provides image information outside the ROI comparable to the FFOV reconstruction. CT using an ROI filter provides a means to reconstruct reliable 3D for a volume of interest with greatly reduced integral dose compared to FFOV projections and with minimal artifacts.

The use of flat panel detectors in computed tomography (CT) systems can improve resolution, reduce system cost, and add operational flexibility by combining fluoroscopy and radiography applications within CT systems. However, some prior studies have suggested that flat panel detectors would not perform well in CT applications due to their lack of high dynamic range, lag artifacts, and inadequate frame rate. The purpose of this study was to perform a physical evaluation of a prototype flat panel detector capable of high frame rates and extended dynamic range. The flat panel detector used had a pixel size of 194 microns and a matrix size of 2048x1536. The detector could be configured for several combinations of frame rate and matrix size up to 750 frames per second for a 512x16 matrix size with 4x4 binning. The evaluation was performed in terms of the MTF and DQE as a function of frame rate and exposure at the IEC RQA5 (~75 kVp, 21 mm Al) beam quality. The image lag was evaluated in terms of temporal-frequency dependent transfer function. Offset shift were also evaluated. Preliminary results indicate 0.1 MTF at 0.92 cycles/mm and DQE(0) of approximately 0.8, 0.6, 0.4, and 0.22 at 0.144, 0.065, 0.035, and 0.008 mR per frame exposures. The temporal MTF exhibited a low-frequency drop and a value of 0.5 at the Nyquist frequency. Offset shift was negligible. Considering high frame rate capabilities of the new detector, the results suggest that the detector has potential for use in real-time CT applications including CT angiography.

We designed, assembled and evaluated a prototype volume CT scanner (VCT) for the purpose of investigating various calibration methods and cone beam reconstruction algorithms as well as the potential clinical benefits of a high-resolution volume CT scanner. The new VCT is based on SIEMENS Sensation4 CT scanner. To achieve larger volume coverage and higher spatial resolution we replaced the prior 4-slices detector with a flat-panel detector. We also modified the prior x-ray tube to achieve a very small focus size by a smaller emitter and wider axial coverage by a larger anode angle. In addition the high-voltage generator was enhanced to support pulsed operation. Special measurement methods were elaborated and applied to measure the focus size, shape and position as well as the uniformity of the flat field x-ray exposure. The accuracy and stability of gantry rotation speed has been evaluated to decide for the most appropriate exposure trigger. New methods are applied to measure and calibrate the resulted x-ray geometry. One prototype VCT scanner is installed at a pre-clinical site to evaluate the application potential of the new VCT technology. The new volume scanner achieves unprecedented spatial resolution, slice sensitivity and spatial coverage. In a complementary paper we present the image quality, contrast resolution and dose issues associated with this scanner.

We investigate cone-beam acquisitions implemented on a novel dedicated cone-beam transmission computed mammotomography (CmT) system with unique arbitrary orbit capability for pendant, uncompressed breasts. We use a previously reported optimized quasi-monochromatic beam technique together with orbits made possible with a novel CmT gantry system, to evaluate Vertical-Axis-Of-Rotation (VAOR), Circle-Plus-Two-Arcs (CP2A), and Saddle trajectories. Aquisition parameters include: W target, 60 kVp tube potential, 100th VL Nd filtration, 1.25 mAs, 55 cm SID, CsI(Tl) digital flat panel x-ray detector, and 7.7cm diameter uniform disc (Defrise) and resolution phantoms. Complex orbits were also performed for a realistic breast phantom. Reconstructions used an iterative ordered subsets transmission (OSTR) algorithm with 4x4 binned projections, 8 subsets, and 10 iterations, with 0.125 mm³ voxels. We evaluate the results for image artifacts, distortion, and resolution. Reconstructed images of the disc coronal and sagittal slices show significant distortion of the discs and phantom interfaces away from the central plane of the cone-beam for VAOR, less distortion for CP2A, and minimal distortion for the complex 3D Saddle orbit. Resolution phantoms indicate no loss of resolution with the Saddle orbit, with the smallest 1.1mm diameter rods clearly resolved. Other image artifacts such as streaking were also significantly reduced in the Saddle orbit case. Results indicate that arbitrary orbits of pendant uncompressed breasts using cone-beam acquisitions and OSTR iterative reconstructions can be successfully implemented for dedicated CmT to improve angular sampling with significant reduction in distortion and other image artifacts. This capability has the potential to improve the performance of dedicated CmT by adequately sampling the breast and anterior chest volumes of prone patients with pendant, uncompressed breasts.

Multi-slice CT today is capable of imaging the heart with excellent temporal resolution. Algorithms have been developed to perform reconstructions combining data from multiple cardiac cycles. This paper presents a simulation phantom that enables a direct measurement of the actual temporal resolution achieved by these algorithms. This is not only useful for assessing the temporal resolution but also for validating the algorithms themselves. A simulation phantom was developed that consists of a 20 cm. diameter water phantom containing an array of cylinders whose intensities are pulsed for various durations ranging from 10 msec. to 250 msec. The intensity varied between the background value of water (0 HU) and 800 HU. By measuring the nominal attenuation value at the center of each cylinder, a curve can be derived representing the response over the given temporal range. A temporal resolution representing the FWHM value is determined based on the half-max value of this curve. Reconstructions were performed using a multi-cycle cardiac algorithm described previously in the literature. The measured FWHM values agree quite well to the temporal resolution predicted by the cardiac algorithm itself. Even the variation along the longitudinal axis can be accounted for by the predicted values. A simulated phantom can be used to accurately assess the temporal resolution of cardiac reconstruction algorithms. Excellent agreement was achieved between the predicted and measured temporal resolution values for the multi-cycle algorithm used in this study.

The sensitivity to detect small breast cancers and the specificity of conventional mammography (CM) remain limited owing to an overlap in the appearances of lesions and surrounding structure. We propose to address the limitations accompanying CM using flat panel detector (FPD)-based cone beam CT breast imaging (CBCTBI). The purpose of the study is to determine optimal x-ray operation ranges for different sizes of normal breasts and corresponding glandular dose levels. The current CBCT prototype consists of a modified GE HighSpeed Advantage CT gantry, an x-ray tube, a Varian PaxScan 4030CB FPD, a CT table and a PC. Two uncompressed breast phantoms, with the diameters of 10.8 and 13.8 cm, consist of three inserts: a layer of silicone jell simulating a background structure, a lucite plate on which five simulated carcinomas are mounted, and a plate on which six calcifications are attached. With a single scan, 300 projections were acquired for all phantom scans. The optimal x-ray techniques for different phantom sizes were determined. The total mean glandular doses for different size phantoms were measured using a CT pencil ionization chamber. With the optimal x-ray techniques that result in the maximal dose efficiency for the different tissue thickness, the image quality with two different phantoms was evaluated. The results demonstrate that the CBCTBI can detect a few millimeter-size simulated carcinoma and ~ 0.2 mm calcification with clinically acceptable mean glandular doses for different size breasts.

There has been considerable interest in megavoltage CT (MVCT) imaging associated with the development of image guided radiation therapy. It is clear that MVCT can provide good image quality for patient setup verification with soft tissue contrast much better than noted in conventional megavoltage portal imaging. In addition, it has been observed that MVCT images exhibit considerably reduced artifacts surrounding metal implants (e.g., surgical clips, hip implants, dental fillings) compared to conventional diagnostic CT images (kVCT). When encountered, these artifacts greatly limit the usefulness of kVCT images, and a variety of solutions have been proposed to remove the artifacts, but these have met with only partial success. In this paper, we investigate the potential for CT imaging in regions surrounding metal implants using high-energy photons from a Cobalt-60 source and from a 4 MV linear accelerator. MVCT and kVCT images of contrast phantoms and a phantom containing a hip prosthesis are compared and analysed. We show that MVCT scans provide good fidelity for CT number quantification in the high-density regions of the images, and in the regions immediately adjacent to the metal implants. They also provide structural details within the high-density inserts and implants. Calculations will show that practical clinical MVCT imaging, able to detect 3% contrast objects, should be achievable with doses of about 2.5cGy. This suggests that MVCT not only has a role in radiotherapy treatment planning and guidance, but may also be indicated for surgical guidance and follow-up in regions where metal implants cannot be avoided.

The integration of 3D-imaging functionality into C-arm systems combines advantages of interventional X-ray systems, e.g. good patient access and live fluoroscopy, with 3D imaging capabilities similar to those of a CT-scanner. To date 3D-imaging with a C-arm system has been mainly used to visualize high contrast objects. However, the advent of high quality flat panel detectors improves the low contrast imaging capabilities. We discuss the influence of scattered radiation, beam hardening, truncated projections, quantization and detector recording levels on the image quality. Subsequently, we present algorithms and methods to correct these effects in order to achieve low contrast resolution. The performance of our pre- and post-reconstructive correction procedures is demonstrated by first clinical cases.

Quantum noise in cone beam CT (CBCT) imaging was studied to provide quantitative relationships among 3D cone beam image noise level and CT acquisition and reconstruction parameters, which include entrance exposure level, number of projections, and single detector size. It showed that the level of reconstructed image noise, which was caused by quantum noise in projection data, was spatially variant and related to the shape of the scan object, and that the image noise level was inversely proportional to the square root of entrance exposure level per projection, square root of number of projections, and square of detector size. Both computer simulations and real phantom studies were conducted to verify the derived quantitative relationships between image noise level and CT parameters. Shepp-logan head phantom was used in computer simulations to verify the theoretical relation between noise level and detector size, while a real cylindrical oil-uniformed phantom was studied to verify the theoretical relation between noise level and entrance exposure level. The real phantom studies were carried out on a flat panel detector (FPD)-based CBCT system available in our Lab. This work can provide a guide on how to balance various CBCT parameters to achieve satisfactory image quality with desired signal-to-noise ratio, specified spatial resolution, low contrast detectability and minimal x-ray radiation to patients.

Cone-beam CT (CBCT) realizes true three-dimensional (3D) imaging in terms of its direct volume reconstruction with isotropic resolution. However, the 3D imaging performance of a CBCT system is spatially variant (or non-uniform) over the support domain, which can be quantitatively characterized by 3D point spread function (PSF). The CBCT system PSF can be experimentally measured through the use of telfon ball and edge-spread technique. For a single circular scan orbit, its volume reconstruction fails to meet data sufficiency condition, consequently causing spatial shift variance. In the pursuit of meeting data sufficiency condition, we have proposed a circle-plus-arc CB scan scheme. The overall CBCT imaging process involves several factors, including x-ray source, cone-beam projection, and computational reconstruction; each factor can in principle be characterized by a convolution kernel or PSF. In this paper, we concentrate on the PSF characterization of circle-plus-arc algorithm. Based on the linearity of Radon transform and inverse Radon transform, we can partition the Radon domain. In particular, the circle-plus-arc scan scheme partitions the Radon domain into a donut-like region (associated with the circle scan) and funnel-like null region (provided by the arc scan). A modified FDK algorithm is responsible for donut region reconstruction, and a filtered-backprojection-styled inverse Radon transform is for the null region reconstruction. By adding them up, we obtain the complete volume reconstruction. Through the use of a bead array phantom (a 5×5×5 array), subject to cone-beam scan under different scan patterns and volume reconstruction, we calculated the local PSF blurs and the spatial variance. The result shows that, for our CBCT simulation with 28 degree cone angle, the circle scan produces a spatial variance of 9.3% and the circle-plus-arc scan reduces that to 1.8%.

Proliferation of coronary stent deployment for treatment of coronary heart disease (CHD) creates a need for imaging-based follow-up examinations to assess patency. Technological improvements in multi-detector computer tomography (MDCT) make it a potential non-invasive alternative to coronary catheterization for evaluation of stent patency; however, image quality with MDCT varies based on the size and composition of the stent. We studied the role of tube focal spot size and power in the optimization of image quality in a stationary phantom. A standard uniform physical phantom with a tubular insert was used where coronary stents (4 mm in diameter) were deployed in a tube filled with contrast to simulate a typical imaging condition observed in clinical practice. We utilized different commercially available stents and scanned them with different tube voltage and current settings (LightSpeed Pro¹⁶, GE Healthcare Technologies, Waukesha, WI, USA). The scanner used different focal spot size depending on the power load and thus allowed us to assess the combined effect of the focal spot size and the power. A radiologist evaluated the resulting images in terms of image quality and artifacts. For all stents, we found that the small focal spot size yielded better image quality and reduced artifacts. In general, higher power capability for the given focal spot size improved the signal-to-noise ratio in the images allowing improved assessment. Our preliminary study in a non-moving phantom suggests that a CT scanner that can deliver the same power on a small focal spot size is better suited to have an optimized scan protocol for reliable stent assessment.

Megavoltage computed tomography (MVCT) has been an active area of research and development in image guided radiation therapy. We have been investigating a particular implementation of MVCT in conjunction with studies of the potential for tomotherapy with a Cobalt-60 radiation source. In this paper, we present results comparing MVCT using a Co-60 source and a 4 MV linear accelerator to conventional kVCT imaging. The Co-60 and linac MVCT measurements were obtained with a first generation benchtop CT imager; the KVCT measurements were obtained using a Philips AcQSim CT Simulator). Phantoms containing various inserts ranging in density from air, through lung, soft tissue and bone equivalent materials and extending to high atomic number metals were imaged with the three modalities. The results enable characterization of image artifacts, CT number linearity and beam hardening. The MVCT images have sufficient contrast that soft tissue regions with 2.8% difference in electron density can be visualized. In MVCT, a linear relationship between CT numbers and electron densities extends to materials with Z ≈ 60. In the 4MV CT imaging there is a position dependence of the CT numbers within a uniform water phantom, which is absent in Co-60 CT images, indicating the presence of beam hardening artifacts in the linac MVCT images. The differences between kVCT and MVCT will be discussed considering the variation of the photon interactions dominating the images. Our investigations indicate that MVCT has properties that may potentially extend its utility beyond radiation therapy.

Low contrast detection tasks, such as the detection of subtle ground glass nodules, are low signal to noise (SNR) situations that can be greatly influenced by choice of reconstruction filter. The goal of this work is to examine tradeoffs in noise, resolution, and dose on the SNR of low contrast test objects, resembling spherical lung nodules, to potentially improve reconstruction filter selection for a given nodule detection task. To perform these experiments, the Modulation Transfer Function (MTF) was calculated for each reconstruction filter available. Next, simulated signal images were created using 2mm section thicknesses of 1 cm diameter spheres of varying contrast levels. The Noise Power Spectra (NPS) were then calculated for each reconstruction filter to be examined. The signal to noise metric used is the ideal Bayesian observer SNR metric, which takes into account the spatial correlations in noise introduced by the filter (and described by the NPS). The IBO SNR was calculated under a variety of reconstruction conditions: (a) varying mAs so that each reconstruction filter results in the same standard deviation; (b) constant mAs and varying reconstruction filter; (c) one reconstruction filter using varying mAs. These measurements provide an opportunity to examine the important tradeoffs in SNR with noise, resolution, and dose that occur with selection of a reconstruction filter and can potentially lead to a quantitative basis for filter selection, improving lesion detectability.

Mega-Voltage systems are used in radiation oncology both for external radiation delivery and patient positioning prior to treatment. A pair of portal images compared with digitally reconstructed radiographs is currently the gold standard for positioning but new developments have made possible the use of Mega-Voltage Cone Beam CT for better 3D setup. The non-ideal imaging geometry of the treatment unit has a direct impact on both methods. It led to the use of a reticule attachment as reference for the scale and the isocenter position on the portal images. The reticule has limited precision and occasionally super-imposes anatomical information. As for Cone Beam, the image quality crucially depends on the knowledge of the scan geometry during the acquisition. The reproducibility of the detector position at each angle will affect the image reconstruction and determine how frequently geometrical calibration must be performed. The objectives of this study are to measure the flex of the detector and evaluate its reproducibility. A RID 1640 Perkin Elmer a-Si Flat Panel is installed on a Siemens Primus linear accelerator with a positioner similar the the one used in the Oncor product. Three original methods are used to investigate the behavior in space and time of the imaging system. A reticule and a Plumb Bob tip are placed along the line formed by the isocenter and the source. Their positions projected on the flat panel for different gantry positions are used to calculate the mechanical flex. Projection matrices obtained in a geometrical Cone Beam calibration are also used to quantify the flat panel sagging. Six full sets of data were acquired over a period of 5 months and recorded overall mechanical flexes of 1 and 3 mm for the transversal and longitudinal directions respectively. The absolute magnitude of the flat panel displacement varies slightly with the method used but the discrepancy stays within the laser precision used for alignment. The small standard deviations of the flat panel displacement (< 1 mm) suggest great stability over time and permits the clinical implementation of patient positioning without the reticule. More experiments on the positioner with the complete set of projection matrices need to be performed to characterize the long-term behavior of the system and to determinate how frequently the Cone Beam calibration needs to be done to conserve image quality. Future work will develop a daily QA protocol to detect possible collisions that would bring the Cone Beam imaging system out of geometrical calibration.

The conventional respiratory-gated CT scan technique includes anatomic motion induced artifacts due to the low temporal resolution. They are a significant source of error in radiotherapy treatment planning for the thorax and upper abdomen. Temporal resolution and image quality are important factors to minimize planning target volume margin due to the respiratory motion. To achieve high temporal resolution and high signal-to-noise ratio, we developed a respiratory gated segment reconstruction algorithm and adapted it to Feldkamp-Davis-Kress algorithm (FDK) with a 256-detector row CT. The 256-detector row CT could scan approximately 100 mm in the cranio-caudal direction with 0.5 mm slice thickness in one rotation. Data acquisition for the RS-FDK relies on the assistance of the respiratory sensing system by a cine scan mode (table remains stationary). We evaluated RS-FDK in phantom study with the 256-detector row CT and compared it with full scan (FS-FDK) and HS-FDK results with regard to volume accuracy and image noise, and finally adapted the RS-FDK to an animal study. The RS-FDK gave a more accurate volume than the others and it had the same signal-to-noise ratio as the FS-FDK. In the animal study, the RS-FDK visualized the clearest edges of the liver and pulmonary vessels of all the algorithms. In conclusion, the RS-FDK algorithm has a capability of high temporal resolution and high signal-to-noise ratio. Therefore it will be useful when combined with new radiotherapy techniques including image guided radiation therapy (IGRT) and 4D radiation therapy.

For the numerical simulation of the radiation transport in the female breast, it is necessary to generate a numerical model of it. Despite of the success achieved by creating such models, for a more precise numerical dosimetry of the radiation transport and optimization of the imaging process, one needs more detailed numerical models. Unfortunately it is not always possible to appropriately accomplish this task on the base of data available today from conventional tomographic devices. The reason is a low contrast in the reconstructed image. This kind of problem was faced when attempting to segment the glandular and interlobular tissues in the 3D data of a breast specimen reconstructed on the Siemens CT device in the University Hospital in Magdeburg. To solve this problem, we have created a set of different 2D x-ray images of the breast specimen on the digital flat panel detector. The 3D distribution of the absorption function of the specimen was then reconstructed from the grey values of these images.

The purpose of our research is to describe the ultimate X-ray detector for angiography. Angiography is a well established X-ray imaging technique for the examination of blood vessels. Contrast agent is injected followed by X-ray exposures and possible obstructions in the blood vessels can be visualized. Standard angiography primarily inspects for possible occlusions and views the vessels as rigid pipes. However, due to the beating heart the flow in arteries is pulsatile. Healthy arteries are not rigid tubes but adapt to various pressure and flow conditions. Our interest is in the (small) response of the artery on the pulse flow. If the arteries responses elastically on the pulse flow, we can expect that it is still healthy. So the detection of artery diameter variations is of interest for the detection of atherosclerosis in an early stage. In this contribution we specify and test a model X-ray detector for its abilities to record the responses of arteries on pulsatile propagating flow distributions. Under normal physiological conditions vessels respond with a temporal increase in arterial internal cross-sectional area of order 10%. This pulse flow propagates along the arteries in response of the left ventricle ejections. We show results of the detection of simulated vessel distensabilities for the model detector and discuss salient parameters features.

Neutron spectroscopy is being developed as a tomographic tool to measure trace element concentration in the body at molecular levels. We are developing a neutron stimulated emission computed tomography (NSECT) system using inelastic scattering of neutrons by target nuclei, to identify elements and their concentration in tissue. An incoming neutron scatters inelastically with an atomic nucleus, which emits a gamma photon of specific energy. This energy, which is detected by an energy-sensitive Gamma detector, is characteristic of the scattering nucleus. The neutron beam and gamma photons undergo considerable attenuation while passing through the body, causing a reduction in detected counts leading to inaccurate reconstruction. We describe a technique to correct for this attenuation as follows. The scanning geometry used for data acquisition is simulated. The lengths of attenuating material lying in the path of the neutron beam are calculated. Neutron attenuation is determined along this path, using attenuation coefficients for each element. Gamma attenuation is calculated similarly for the path between the point of gamma origin and the detector. A transmission profile is then determined for each projection, using the product of the neutron and gamma attenuations for every point along the projection. The inverse of the integral of this profile yields a correction factor. The experimental data is multiplied by the correction factors to yield attenuation corrected projections. After correction, the projection data is seen to represent the known elemental distribution more accurately. This correction technique improves the consistency of the projections, and leads to the improved accuracy in reconstructed NSECT images.

We have developed a patient dosimetry tool that will permit the optimization of image quality in CT. Published Monte Carlo CT dosimetry data were used to generate values of the energy imparted to an anthropomorphic phantom undergoing head and body CT examinations. Energy imparted factors E _5,n were computed for irradiation of 5 mm thick slabs of the anthropomorphic phantom, which were normalized to the free-in-air dose to muscle at the CT isocenter obtained in the absence of any phantom or patient. Calculations of E _5,n were obtained for CT scanners from five vendors and for 208 contiguous slabs of the anthropomorphic phantom ranging from the top of the head to the upper leg region. Values of E _5,n were relatively constant in the abdomen and chest region, but there was a large inter-scanner variability, with a mean of 170 ± 50 mJ/Gy for five vendors when E _5,n values were averaged over the whole trunk. The mean E _5,n for the neck region was 100 ± 20 mJ/Gy, which increased to 110 ± 30 mJ/Gy for the head region. Adding 0.25 mm Cu filtration increased the value of the normalized energy imparted value E _5,n by an average of 24%. Relative to 100 kVp, increasing the x-ray tube voltage by 20 and 30 kVp increased the normalized energy imparted value E _5,n by 10% and 14%, respectively. Patient energy imparted is useful for studying optimization strategies with respect to x-ray technique factors.

Monochromatic beams produced with synchrotron sources are known to give higher contrast for mammography than clinical broadband sources. Monochromatic beams could also be achieved with clinical x-ray sources, by diffraction off of flat monochromator crystals, but monochromatic intensities are too low for imaging because only a small fraction of the incident beam is at the right energy and angle. With the use of polycapillary optics, monochromatic intensities could be increased. Two different x-ray optics schemes were tested to provide high monochromatic intensity from conventional divergent sources. A polycapillary collimating optic was employed to collect a large solid angle and redirect it into a parallel beam, which can be efficiently diffracted from a flat crystal. Measurements were performed for crystals of varying angular acceptance because there is a trade-off between intensity and resolution. Alternatively, doubly curved crystal (DCC) optics can be used to collect and focus monochromatic x rays from a divergence source. Higher monochromatic intensity can be obtained because the DCC optic diffracts and focuses the incident beam across the whole area of the crystal. For both methods, monochromatization occurs before the patient, resulting in a potential dose reduction as well as significant measured contrast enhancement. Measurements were made of contrast, resolution and intensity for the two techniques, and were compared to each other and to theoretical calculations.

In existing proton treatment centers, dose calculations are performed based on x-ray computerized tomography (CT). Alternatively, the therapeutic proton beam could be used to collect the data for treatment planning via proton CT (pCT). With the development of medical proton gantries, first at Loma Linda University Medical Center and now in several other proton treatment centers, it is of interest to continue the early pCT investigations of the 1970s and the early 1980s. From that time, the basic idea of the pCT method has advanced from average energy loss measurements to an individual proton tracking technique. This reduces the image degradation due to multiple Coulomb scattering. Thereby, the central pCT problem shifts to the fidelity of the physical information obtained about the scanned patient, which will be used for proton treatment planning. The accuracy of relative electron density distributions extracted from pCT images was investigated in this work using continuous slowing down approximation (CSDA) and water-equivalent-thickness (WET) concepts. Analytical results were checked against Monte Carlo simulations, which were obtained with SRIM2003 and GEANT4 Monte Carlo software packages. The range of applications and the sources of absolute errors are discussed.

Lung disease is the third leading cause of death in the western world. Lung air volume measurements are thought to be early indicators of lung disease and markers in pharmaceutical research. The purpose of this work is to develop a lung phantom for assessing and comparing the quantitative accuracy of hyperpolarized xenon 129 magnetic resonance imaging (HP ¹²⁹Xe MRI), conventional computed tomography (HRCT), and highresolution small-animal CT (μCT) in measuring lung gas volumes. We developed a lung phantom consisting of solid cellulose acetate spheres (1, 2, 3, 4 and 5 mm diameter) uniformly packed in circulated air or HP ¹²⁹Xe gas. Air volume is estimated based on simple thresholding algorithm. Truth is calculated from the sphere diameters and validated using μCT. While this phantom is not anthropomorphic, it enables us to directly measure air space volume and compare these imaging methods as a function of sphere diameter for the first time. HP ¹²⁹Xe MRI requires partial volume analysis to distinguish regions with and without ¹²⁹Xe gas and results are within %5 of truth but settling of the heavy ¹²⁹Xe gas complicates this analysis. Conventional CT demonstrated partial-volume artifacts for the 1mm spheres. μCT gives the most accurate air-volume results. Conventional CT and HP ¹²⁹Xe MRI give similar results although non-uniform densities of ¹²⁹Xe require more sophisticated algorithms than simple thresholding. The threshold required to give the true air volume in both HRCT and μCT, varies with sphere diameters calling into question the validity of thresholding method.

PURPOSE: Rheumatiod arthritis (RA) of the hand is a widespread and debilitating disease with a large social and economic impact. RA damages the inter articular cartilage, causing narrowing of the joint space width (JSW.) In this study a template-based approach for measuring the 3D JSW is presented, which uses multiple projections of the joint and a comparison to 3D templates from anatomical specimens. METHODS: This study examined 20 proximal interphalangeal (PIP), and 20 metacarpophalangeal (MCP) joints. Realistic simulated radiographs with a pitch of 0.1 mm were produced from 3D data sets obtained by imaging skeletal specimens with a 0.066 mm pitch μCT. The technique attempted to match each bone to an anatomical template after the template had undergone transformations in virtual imaging space.For each native joint projection, the two opposing margins were delineated using a technique developed to segment bone margins on hand radiographs. The projected margins of each joint surface at each acquisition were then compared to a set of margins created by projecting rays through the surfaces of the anatomical templates, and the best-matched template was chosen. The performance was evaluated by calculating dJSW = |Native joint JSW- Template JSW|. RESULTS: The average dJSW ranged from 0.10 mm to 0.30 mm for approximately half of the average joint radius.CONCLUSIONS: A new technique for 3D imaging based on anatomical templates has been demonstrated. The work is specific to hand imaging but may be applied to other anatomical objects with regularity of shape. It may also complement previous work in digital tomosynthesis.

The determination of optical properties of a semi-infinite medium such as biological tissue has been widely investigated by many authors. Reflectance formulas can be derived from the diffusion equation for different boundary conditions at the medium-air interface. This quantity can be measured at the medium surface. For realistic objects, such as a mouse, tissue optical properties can only be determined at the object surface. However, near the surface, the diffusion approximation is weak and boundary models have to be considered. In order to investigate the validity of the time resolved reflectance approach at the object boundary, we have estimated optical properties of a liquid semi-infinite medium by this method for different boundary conditions and different positions of the fibers beneath the surface.The time-correlated single photon counting (TCSPC) technique is used to measure the reflectance curve. Our liquid phantoms are made of water, white paint and Ink. Laser light is delivered by a pulsed laser diode. Measurements are then fitted to theoretical solutions expressed as a function of source and detector’s depths and distance.By taking as reference the optical properties obtained from the infinite model for fibers deeply immersed, the influence of the different boundary conditions and bias induced are established for different fibers' depths and a variety of solutions. This influence is analyzed by comparing evolution of the reflectance models, as well as estimations of absorption and reduced scattering coefficients.

Cervical cancer is the second most common cancer among women worldwide. Early detection of cervical cancer is very important for successful treatment and increasing survival. Papanicolaou test is the most popular and effective screening test for cervical cancer, but it is highly subjective and skilled-labor intensive. We report a multispectral imaging microscopic system for Papanicolaou smear analysis for early detection of cervical cancer. Different from traditional color imaging method, we use multispectral imaging techniques for image acquisition, which can simultaneously record spectral and spatial information of a sample. A liquid crystal tunable filter (LCTF) is coupled in the light microscope for fast wavelength selection and a two-dimensional cooled charge-coupled device (CCD) for image capture. In this paper, the multispectral image acquisition method is introduced, including exposure control and spectral calibration, which makes the images not so dependent on imaging devices. In the image segmentation process, an effective algorithm using spectral ratio method is applied for cell nuclei detection. This segmentation method can easily detect the nuclei and diminish the influence of the cytoplasm overlap. Results show that our segmentation is more robust and precise than conventional color imaging method which is heavily dependant on sample staining and illumination conditions while with high speed. Once the nuclei have been segmented, cell features including morphological and textural features are measured. A genetic algorithm is used for feature selection and a support vector machine(SVM)is used for training and classification. This paper is focused on image acquisition and segmentation.

The tantalum plasma flash x-ray generator is useful in order to perform high-speed K-edge angiography using cone beams because Kα rays from the tantalum target are absorbed effectively by gadolinium-based contrast media. In the flash x-ray generator, a 150 nF condenser is charged up to 80 kV by a power supply, and flash x rays are produced by the discharging. The x-ray tube is a demountable diode, and the turbomolecular pump evacuates air from the tube with a pressure of approximately 1 mPa. Since the electric circuit of the high-voltage pulse generator employs a cable transmission line, the high-voltage pulse generator produces twice the potential of the condenser charging voltage. When the charging voltage was increased, the K-series characteristic x-ray intensities of tantalum increased. The K lines were clean and intense, and hardly any bremsstrahlung rays were detected. The x-ray pulse widths were approximately 100 ns, and the time-integrated x-ray intensity had a value of approximately 300 μGy at 1.0 m from the x-ray source with a charging voltage of 80 kV. Angiography was performed using a film-less computed radiography (CR) system and gadolinium-based contrast media. In angiography of non-living animals, we observed fine blood vessels of approximately 100 μm with high contrasts.

Cone beam CT mammography (CBCTM) is an emerging breast imaging technology and is currently under intensive investigation [1-3]. One of the major challenges in CBCTM is to understand the characteristics of scatter radiation and to find ways to reduce or correct its degrading effects. Since the breast shape, geometry and image formation process are significantly different from conventional mammography, all system components and parameters such as target/filter combination, kVp range, source to image distance, detector design etc. should be examined and optimized. In optimizing CBCTM systems, it is important to have knowledge of how different imaging parameters affect the recorded scatter within the image. In this study, a GEANT4 based Monte Carlo simulation package (GATE) was used to investigate the scatter magnitude and its’ distribution in CBCTM. The influences of different air gaps, kVp settings, breast sizes and breast composition on the scatter primary ratio (SPR) and scatter profiles were examined. In general, the scatter to primary ratio (SPR) is strongly dependent on the breast size and air gap, and is only moderately dependent on the kVp setting and breast composition. These results may be used for optimization of CBCTM systems, as well as for developing scatter correction methods.

We validate the computer-based simulation tools developed in our laboratory for use in high-resolution CT research. The 4D NURBS-based cardiac-torso (NCAT) phantom was developed to provide a realistic and flexible model of the human anatomy and physiology. Unlike current phantoms in CT, the 4D NCAT has the advantage, due to its design, that its organ shapes can be changed to realistically model anatomical variations and patient motion. To efficiently simulate high-resolution CT images, we developed a unique analytic projection algorithm (including scatter and quantum noise) to accurately calculate projections directly from the surface definition of the phantom given parameters defining the CT scanner and geometry. The projection data are reconstructed into CT images using algorithms developed in our laboratory. The 4D NCAT phantom contains a level of detail that is close to impossible to produce in a physical test object. We, therefore, validate our CT simulation tools and methods through a series of direct comparisons with data obtained experimentally using existing, simple physical phantoms at different doses and using different x-ray energy spectra. In each case, the first-order simulations were found to produce comparable results (<12%). We reason that since the simulations produced equivalent results using simple test objects, they should be able to do the same in more anatomically realistic conditions. We conclude that, with the ability to provide realistic simulated CT image data close to that from actual patients, the simulation tools developed in this work will have applications in a broad range of CT imaging research.

The use of focused anti-scatter grids on digital radiographic systems with two-dimensional detectors produces acquisitions with a decreased scatter to primary ratio and thus improved contrast and resolution. Simulation software is of great interest in optimizing grid configuration according to a specific application. Classical simulators are based on complete detailed geometric descriptions of the grid. They are accurate but very time consuming since they use Monte Carlo code to simulate scatter within the high-frequency geometric description of the grid. We propose a new practical method which couples an analytical simulation of the grid interaction with a radiographic system simulation program. First, a two dimensional matrix of probability depending on the grid is created offline, in which the first dimension represents the angle of impact with respect to the normal to the grid lines and the other the energy of the photon. This matrix of probability is then used by the Monte Carlo simulation software in order to provide the final x-rays scatter flux image. To evaluate the gain of CPU time, we define the increasing factor as the increase of CPU time of the simulation with as or without the grid. Increasing factors were calculated with the new model and with classical methods representing the grid with a Computed-Aided Designed (CAD) model. With this new method, increasing factors are shortened by three to four orders of magnitude.

In recent years, there has been interest in exploring the feasibility of CT breast imaging using flat-panel digital detectors in a truncated cone-beam geometry. Preliminary results are promising and it appears as if 3D tomographic imaging of the breast has great potential for reducing the masking effect of superimposed parenchymal structure typically observed with conventional mammography. In this study, a mathematical framework used for determining optimal design and acquisition parameters for such a CT breast imaging system is described. The ideal observer SNR is used as a figure-of-merit, under the assumptions that the imaging system is linear and shift-invariant. Computation of the ideal observer SNR used a parallel-cascade model to predict signal and noise propagation through the detector, as well as a realistic model of the lesion detection task in breast imaging. For all optimizations discussed here, the total mean glandular dose for a CT breast imaging study is constrained to be approximately equivalent to that of a two-view conventional mammography study. The framework presented is used to explore the affect of the specific task on the optimal exposure technique of flat-panel CT breast imaging. In particular, it is observed that modeling the normal mammographic structure in the projection images can sometimes impact the optimal kVp settings.

Although conventional mammography is currently the best modality to detect early breast cancer, it is limited in that the recorded image represents the superposition of a 3D object onto a 2D plane. As an alternative, cone-beam CT breast imaging with a CsI based flat-panel imager (CTBI) has been proposed with the ability to provide 3D visualization of breast tissue. To investigate possible improvements in lesion detection accuracy using CTBI over digital mammography (DM), a computer simulation study was conducted using simulated lesions embedded into a structured 3D breast model. The computer simulation realistically modeled x-ray transport through a breast model, as well as the signal and noise propagation through the flat-panel imager. Polyenergetic x-ray spectra of W/Al 50 kVp for CTBI and Mo/Mo 28 kVp for DM were modeled. For the CTBI simulation, the intensity of the x-ray spectra for each projection view was determined so as to provide a total mean glandular dose (MGD) of 4 mGy, which is approximately equivalent to that given in a conventional two-view screening mammography study. Since only one DM view was investigated here, the intensity of the DM x-ray spectra was defined to give 2 mGy MGD. Irregular lesions were simulated by using a stochastic growth algorithm providing lesions with an effective diameter of 5 mm. Breast tissue was simulated by generating an ensemble of backgrounds with a power law spectrum. To evaluate lesion detection accuracy, a receiver operating characteristic (ROC) study was performed with 4 observers reading an ensemble of images for each case. The average area under the ROC curves (A_z) was 0.94 for CTBI, and 0.81 for DM. Results indicate that a 5 mm lesion embedded in a structured breast phantom can be detected by CT breast imaging with statistically significant higher confidence than with digital mammography.

We describe MANTIS (Monte carlo x-rAy electroN opTical Imaging Simulation), a tool for simulating imaging systems that tracks x rays, electrons, and optical photons in the same geometric model. The x-ray and electron transport and involved physics models are from the PENELOPE package and include elastic and inelastic scattering, and bremsstrahlung from 100 eV to 1 GeV. The optical transport and corresponding physics models are from DETECT-II and include Fresnel refraction and reflection at material boundaries, bulk absorption and scattering. X rays are generated using the flexible source description from PENELOPE. When x rays or electrons interact and deposit energy in the scintillator, the code generates a number of optical quanta at that location, according to a model for the conversion process. The optical photons are then tracked until they reach an absorption event that in some cases contributes to the electronic signal. We demonstrate the capabilities of the new tool with respect to x-ray source, object to be imaged, and detector models. Of particular importance is the improved geometric description of structured phosphors that can handle tilted columns in needle-like phosphor screens. Examples of the simulation output with respect to signal blur and pulse-height distributions of the scintillation light are discussed and compared with previously published experimental results.

Contrast enhanced digital mammography (CEDM), which is based upon the analysis of a series of x-ray projection images acquired before/after the administration of contrast agents, may provide physicians critical physiologic and morphologic information of breast lesions to determine the malignancy of lesions. This paper proposes to combine the kinetic analysis (KA) of contrast agent uptake/washout process and the dual-energy (DE) contrast enhancement together to formulate a hybrid contrast enhanced breast-imaging framework. The quantitative characteristics of materials and imaging components in the x-ray imaging chain, including x-ray tube (tungsten) spectrum, filter, breast tissues/lesions, contrast agents (non-ionized iodine solution), and selenium detector, were systematically modeled. The contrast-noise-ration (CNR) of iodinated lesions and mean absorbed glandular dose were estimated mathematically. The x-ray techniques optimization was conducted through a series of computer simulations to find the optimal tube voltage, filter thickness, and exposure levels for various breast thicknesses, breast density, and detectable contrast agent concentration levels in terms of detection efficiency (CNR²/dose). A phantom study was performed on a modified Selenia full field digital mammography system to verify the simulated results. The dose level was comparable to the dose in diagnostic mode (less than 4 mGy for an average 4.2 cm compressed breast). The results from the computer simulations and phantom study are being used to optimize an ongoing clinical study.

The performance of an inverse geometry volumetric CT (IGCT) system with multiple detector arrays is being investigated. The system is capable of a complete acquisition of a volume free from cone-beam artifacts with only a single rotation of the gantry. The IGCT system is composed of a large source array opposite three small detector arrays with a field-of-view (FOV) large enough for clinical imaging (45cm). Simulations were conducted to estimate the MTF at different points in the FOV. The simulations involved generating 2D projection data of a 100um circular object followed by a reconstruction algorithm that uses gridding and filtered backprojection. The simulations also modeled finite source spot and detector element sizes. The estimated MTF’s were compared with theoretical MTF’s at 0 cm, 10 cm, and 20 cm away from the isocenter. The simulated MTF’s closely matched the theoretical MTF’s. The MTF in the radial direction was over 10% at 16 lp/cm across the entire FOV while the azimuthal MTF 10% point degraded to 10.4 lp/cm at the edge of the FOV. This degradation in azimuth, which can be corrected for, is due to gridding in the angular direction which is magnified at large distances away from the isocenter. The simulations show promising results for the in-plane resolution of the multiple detector array IGCT system. Noise properties and other factors impacting performance are currently being investigated.

A method is presented for accurately simulating the effects of dose reduction in x-ray computed tomography (CT) by adding synthetic noise to raw projection data. A model for realistic noise in projection data was generated, incorporating the mechanisms of stochastic noise in energy-integrating x-ray detectors, electronic system noise, and bowtie beam filtering (used for patient dose reduction). Parameters for the model were extracted from phantom measurements on a variety of clinical CT scanners (helical single row, four-row, and 16-row). Dose reduction simulations were performed by adding synthetic noise based on the noise model to raw data acquired from clinical scanners. Qualitative and quantitative validation of the process was accomplished by comparing phantom scans acquired under high and low dose conditions with simulated imagery. The importance of including alternative noise mechanisms (bowtie filter and system noise) was demonstrated. Henceforth, scans of clinical patients were acquired using conventional protocols; through simulations, image sets were presented at a variety of lower dose procedures. The methodology promises to be a useful tool for radiologists to explore dose reduction protocols in an effort to produce diagnostic images with radiation dose “as low as reasonably achievable”.

We have developed a computer simulation model for cone beam computed tomography (CT) chest imagingon a general-purpose personal computer cluster system. Our simulation model incorporates quantum noise, detector blurring, and additive system noise.The main objective is to study how x-ray dose would affect the detectabilityof nodules in simulated cone beam CT chest images. The Radon transforms formalism was used to calculate the projection views for an analytically modeled chest phantom. A parallel random number generator was then used to simulate and add quantum noise whose level depends on the incident x-ray fluence, detector quantum efficiency and pixel size (0.4 mm).We also simulated detector blurring by convolving the noise added images with a Gaussian function matching the modulation transfer function measured for the flat panel x-ray detector studied. Then we modeled the additive system noise and added to the final projection images.The noise level (σ=20) for the additive system noise was calculated from the noise power spectrum of the flat panel detector using the curve-fitting technique.The Feldkamp algorithm with a Gaussian pre-filtering processwas used to reconstruct 3D image data from the projection images.For nodule contrast, the linear attenuation coefficient difference between nodule and lung was set to 10.0%. The diameters for the spherical nodules ranged from 0.2 to 1.7 cm. It was found that our Gaussian pre-filtering process helped reduce the noise level in the reconstructed images and allowed the nodules to be better visualized significantly. At 100,000 photons per pixel (8000 mR total unattenuated exposure at the rotating center), nodules 0.3 mm or larger could be visualized; at 10,000 photons per pixel( 800 mR), nodules 0.5 mm or larger could be visualized; at 2000 photons per pixel (160 mR), only nodules 1.5 mm or larger could be visualized.

We present a simulation study of the countrate performance of a PET scanner with partial collimation. In this study, partial collimation is achieved by removal of every other septum from the standard 2D septa set for the GE Advance PET scanner. System behavior is evaluated with a photon tracking simulation package (SimSET) and calibrated to measured data for 2D and fully-3D acquisition modes using the NEMA NU-2 countrate phantom. Results are evaluated in terms of true, scattered, and random coincidences and noise equivalent counts (NEC) regarding both counts per image plane and total counts as a function of activity. Our results show a good agreement between the measured and simulated count rates for the Advance PET scanner for the 2D (full collimation) and fully-3D (no collimation) acquisition modes, increasing our confidence in the predicted countrate results for the partial collimation mode. The latter results in a countrate performance intermediate between the 2D and fully-3D acquisition modes and yields a more favorable countrate performance for clinical activity levels.

A number of different technologies are available for digital mammography. However, it is not clear how differences in the physical performance aspects of the different imaging technologies affect clinical performance. Randomised controlled trials provide a means of gaining information on clinical performance however do not provide direct comparison of the different digital imaging technologies. This work describes a method of simulating the performance of different digital mammography systems. The method involves modifying the imaging performance parameters of images from a small field of view (SFDM), high resolution digital imaging system used for spot imaging. Under normal operating conditions this system produces images with higher signal-to-noise ratio (SNR) over a wide spatial frequency range than current full field digital mammography (FFDM) systems. The SFDM images can be 'degraded’ by computer processing to simulate the characteristics of a FFDM system. Initial work characterised the physical performance (MTF, NPS) of the SFDM detector and developed a model and method for simulating signal transfer and noise properties of a FFDM system. It was found that the SNR properties of the simulated FFDM images were very similar to those measured from an actual FFDM system verifying the methodology used. The application of this technique to clinical images from the small field system will allow the clinical performance of different FFDM systems to be simulated and directly compared using the same clinical image datasets.

This paper investigates the feasibility of using a flat panel based cone-beam computer tomography (CT) system for 3-D breast imaging with computer simulation and imaging experiments. In our simulation study, 3-D phantoms were analytically modeled to simulate a breast loosely compressed into cylindrical shape with embedded soft tissue masses and calcifications. Attenuation coefficients were estimated to represent various types of breast tissue, soft tissue masses and calcifications to generate realistic image signal and contrast. Projection images were computed to incorporate x-ray attenuation, geometric magnification, x-ray detection, detector blurring, image pixelization and digitization. Based on the two-views mammography comparable dose level on the central axis of the phantom (also the rotation axis), x-ray kVp/filtration, transmittance through the phantom, detected quantum efficiency (DQE), exposure level, and imaging geometry, the photon fluence was estimated and used to estimate the phantom noise level on a pixel-by-pixel basis. This estimated noise level was then used with the random number generator to produce and add a fluctuation component to the noiseless transmitted image signal. The noise carrying projection images were then convolved with a Gaussian-like kernel, computed from measured 1-D line spread function (LSF) to simulated detector blurring. Additional 2-D Gaussian-like kernel is designed to suppress the noise fluctuation that inherently originates from projection images so that the reconstructed image detectability of low contrast masses phantom can be improved. Image reconstruction was performed using the Feldkamp algorithm. All simulations were performed on a 24 PC (2.4 GHz Dual-Xeon CPU) cluster with MPI parallel programming. With 600 mrads mean glandular dose (MGD) at the phantom center, soft tissue masses as small as 1 mm in diameter can be detected in a 10 cm diameter 50% glandular 50% adipose or fatter breast tissue, and 2 mm or larger masses are visible in a 100% glandular 0% adipose breast tissue. We also found that the 0.15 mm calcification can be detected for 100μm detector while only 0.2 μm or above are visible for 200 μm detector. Our simulation study has shown that the cone-beam CT breast imaging can provide reasonable good quality and detectability at a dose level similar to that of two views\mammography. For imaging experiments, a stationary x-ray source and detector, a step motor driven rotating phantom system was constructed to demonstrate cone beam breast CT image. A breast specimen from mastectomy and animal tissue embedded with calcifications were imaged. The resulting images show that 355-425 μm calcifications were visible in images obtained at 77 kVp with a voxel size of 316 μm and a center dose of 600 mrads. 300-315 μm calcifications were visible in images obtained at 60 kVp with a voxel size of 158 μm and a center dose of 3.6 rads.

We have designed a detector based on a-Se(amorphous Selenium) and done simulation the detector with Monte Carlo method. We will apply the cascaded linear system theory to determine the MTF for whole detector system. For direct comparison with experiment, we have simulated 139um pixel pitch and used simulated X-ray tube spectrum.

A dominant component of image quality for whole-body positron emission tomography (PET) imaging is attenuation, which is determined by patient thickness. This can be partially compensated for by adjusting scan duration. We evaluate the effect of changes in patient thickness and scan duration on lesion detection with model observers. We simulated 2D PET acquisitions of an anthropomorphic phantom with spherical target lesions. Three different anthropomorphic phantoms were used, with effective abdominal diameters of 20 cm, 27 cm, and 35 cm. The diameters of the lesions were varied from 1.0 to 3.0 cm, and the contrast ratios of the lesions were varied from 1.5 to 4.0. Noise-free scans were simulated with an analytical simulator. Poisson noise was added to simulate scan durations ranging from 1 to 10 minutes per bed position, using noise equivalent count rates previously measured using a modified NEMA NU2 countrate phantom. The average detectability of each target lesion under each condition was calculated using a non-prewhitening matched filter from 25 noisy realizations for each combination of parameters. Our results demonstrate the variation of the minimum scan duration required to detect a target of a given size and contrast ratio, for any fixed threshold of detectability. For image quality to remain constant for patients with larger cross-sectional areas, acquisition times should be increased accordingly, although in some cases this may not be possible due to practical constraints.

A wavelet based method of noise reduction has been tested for mammography using computer-simulated images for which the truth is known exactly. This method is based on comparing two images at different scales, using a cross-correlation-function as a measure of similarity to define the image modifications in the wavelet domain. The computer-simulated images were calculated for noise-free primary radiation using a quasi-realistic voxel phantom. Two images corresponding to slightly different geometry were produced. Gaussian noise was added with a mean value of zero and a standard deviation equal to 0.25% to 10% of the actual pixel value to simulate quantum noise with a certain level. The added noise could be reduced by more than 70% using the proposed method without any noticeable corruption of the structures for 4% added noise. The results indicate that it is possible to save 50% dose in mammography by producing two images (each 25% of the dose for a standard mammogram). Additionally, a reduction or even a removal of the anatomical noise might be possible and therefore better detection rates of breast cancer in mammography might be achievable.

We have developed an adaptive filtering algorithm for cardiac CT scans with EKG-modulated tube current to optimize resolution and noise for different cardiac phases and to provide safety net for cases where end-systole phase is used for coronary imaging. This algorithm has been evaluated using patient cardiac CT scans where lower tube currents are used for the systolic phases. In this paper, we present the evaluation results. The results demonstrated that with the use of the proposed algorithm, we could improve image quality for all cardiac phases, while providing greater noise and streak artifact reduction for systole phases where lower CT dose were used.

The new multi slice CT scanners are characterized by a wide coverage (collimation) in the longitudinal (z) direction. This large collimation provides large and rapid coverage, it may however introduce artifacts caused by scatter, which are enlarged with collimation. The scattering in the z direction induces artifacts that appear as dark shadows between bones.We present a software correction that reduces the scattering artifacts.The correction is based on a subtraction of a low frequency offset, as foreseen in simulations of scatter effect, from the raw data. The study includes tests of the scattering correction of both phantoms and clinical scans.The corrected images are compared to un-corrected images scanned with the same z-coverage (collimation) and to images scanned with narrower z-coverage with the same scan parameters. The results demonstrate that the correction of the images, scanned with the wide collimation, reduces the scattering artifact significantly. The obtained image quality levels to that of the images scanned with narrower collimation.

We have developed a statistically principled sinogram smoothing approach for single-slice, multi-slice, and conebeam computed tomography (CT) with the aim of obtaining high-quality reconstructed images from scans conducted with lower radiation doses than are generally employed. Reducing patient dose in x-ray computed tomography (CT) can be achieved by reducing the output of the x-ray tube but this will generally increase image noise levels as well. Standard approaches to noise control in CT involve simple apodization of the reconstruction filter, which would unacceptably compromise resolution to achieve the noise control necessary in low-dose situations. In this work, we present an explicitly statistical approach to sinogram smoothing in which we formulate the estimation of the attenuation line integrals from the measured data as a statistical estimation problem and make use of penalized likelihood methods to perform the estimation. We find that the proposed penalized-likelihood sinogram smoothing approach can reduce the appearance of noise in reconstructed images without introducing additional artifacts or severely degrading resolution. In terms of resolution-noise tradeoffs, it was found to outperform other approaches considered.

Nowadays, most of treatments for external radiotherapy are prepared with Treatment Planning Systems (TPS) which uses a virtual patient generated by a set of transverse slices acquired with a CT scanner of the patient in treatment position ^{1 2 3}. In the first step of virtual simulation, the TPS is used to define a ballistic allowing a good target covering and the lowest irradiation for normal tissues. This parameters optimisation of the treatment with the TPS is realised with particular graphic tools allowing to: •Contour the target, •Expand the limit of the target in order to take into account contouring uncertainties, patient set up errors, movements of the target during the treatment (internal movement of the target and external movement of the patient), and beam's penumbra, •Determine beams orientation and define dimensions and forms of the beams, •Visualize beams on the patient's skin and calculate some characteristic points which will be tattooed on the patient to assist the patient set up before treating, •Calculate for each beam a Digital Reconstructed Radiography (DRR) consisting in projecting the 3D CT virtual patient and beam limits with a cone beam geometry onto a plane. These DRR allow one for insuring the patient positioning during the treatment, essentially bone structures alignment by comparison with real radiography realized with the treatment X-ray source in the same geometric conditions (portal imaging). Then DRR are preponderant to insure the geometric accuracy of the treatment. For this reason quality control of its computation is mandatory⁴ . Until now, this control is realised with real test objects including some special inclusions^{4 5} . This paper proposes to use some numerical test objects to control the quality DRR calculation in terms of computation time, beam angle, divergence and magnification precision, spatial and contrast resolutions. The main advantage of this proposed method is to avoid a real test object CT acquisition allowing for a drastic time reduction of the control as well as its automatic control. This method has been used to test a new method to compute DRR⁶ and is here presented to control a standard DRR calculation algorithm⁷ .

A novel exact fan-beam image reconstruction formula is presented and validated using both mathematical phantom data and clinical data. This algorithm takes the form of the standard ramp filtered backprojection (FBP) algorithm plus local compensation terms. An equal weighting scheme is utilized in this algorithm in order to properly account for redundantly measured projection data. The algorithm has the desirable property of maintaining a mathematically exact result for: the full scan mode (2π), the short scan mode (π+ full fan angle), and the super-short scan mode (less than (π + full fan angle)). Another desirable feature of this algorithm is that it is derivative-free. The derivative-free nature of this algorithm distinguishes it from other exact fan-beam FBP algorithms.

For non-uniformity correction a flat field x-ray image is needed, and to obtain it the center of detector is usually aligned with the focal spot of the x-ray tube, which is conserved when examining patients to preserve the flat field. In some of radiographic techniques, however, it is necessary to move the x-ray tube off the center position of detector or tilt the detector. We investigated the effect of X-ray tube positions with respect to detector on the non-uniformity correction, and propose a method to reduce the effect using a new algorithm with computer simulation. Gain images were taken in two SIDs. Pixel values at second SID was calculated using the pixel values at first SID, gain coefficient that represents pixels own unique radiation sensitivity characteristics and the formula based on the solid angle of each detector pixel facing to the x-ray source. Gain coefficient was adjusted using the difference between calculated and real pixel values. Calculation was repeated with new gain coefficient until the gain coefficient was converged into prescribed range. Non-uniformity of blank x-ray images taken with the detector tilted by 0 to 45 degrees was corrected and five ROIs across the image were defined and analyzed. When the proposed algorithm was used for the flat field correction standard deviations of pixel values in the ROIs were reduce to 10% of the cases of usual flat field correction.

A modified Feldkamp (FDK) half-scan (MFDKHS) algorithm was developed to conduct the circular half-scan scheme on a flat panel detector (FPD)-based CBCT. X-ray source scans the object in a circular trajectory in the range of 180 degrees plus full fan angle rather than 360 degrees for a full scan. The redundant data is weighted by a weighting function in 3-D case based on the ones proposed by Parker¹, FDK algorithm is then applied on this weighted data, and another FBP term developed by Dr. Hu is also added to get the final reconstruction. The MFDKHS is expected to correct the attenuation coefficient drop of the reconstructed object associated with FDK in the place where Z (rotation axis) is away from scanning plane and improve the temporal resolution as well.

Image registration is an important problem for image processing and computer vision with many proposed applications in medical image analysis.^{1, 2} Image registration techniques attempt to map corresponding features between two images. The problem is particularly difficult as anatomy is subject to elastic deformations. This paper considers this problem in the context of graph matching. Firstly, weighted Region Adjacency Graphs (RAGs) are constructed from each image using an approach based on watershed saliency. ³ The vertices of the RAG represent salient regions in the image and the (weighted) edges represent the relationship (bonding) between each region. Correspondences between images are then determined using a weighted graph matching method. Graph matching is considered to be one of the most complex problems in computer vision, due to its combinatorial nature. Our approach uses a multi-spectral technique to graph matching first proposed by Umeyama⁴ to find an approximate solution to the weighted graph matching problem (WGMP) based on the singular value decomposition of the adjacency matrix. Results show the technique is successful in co-registering 2-D MRI images and the method could be useful in co-registering 3-D volumetric data (e.g. CT, MRI, SPECT, PET etc.).

We investigated an image reconstruction algorithm to reduce cone-beam artifacts in cone-beam CT. To examine the factors involved in the occurrence of cone-beam artifacts, micro-spheres phantom were arranged longitudinally at different positions and a computer simulation was performed. Due to differences in projection angle, data projected onto the detector surface were projected along trajectories shown as different periodic functions depending on the distance and position from the center of rotation. Therefore, projection along several detector channels based on different projection data resulting from different periodic functions is considered responsible for the increase in cone-beam artifacts associated with an increase in the distance of reconstruction planes from the center of rotation. Our new algorithm to reduce such artifacts features: 1) A change in weighting with respect to projection data obtained at different projection angles. 2) Distribution of correction coefficients so that they are larger near the center of the detector, while taking individual channel data for the detector into account, and smaller near the edges. 3) Three-dimensional back-projection of corrected projection data. The effect of the reduction in cone-beam artifacts of an object located at the edges markedly enhanced reconstruction planes at positions further from the center of rotation.

Aging and smoking history increases number of pulmonary emphysema. Alveoli restoration destroyed by pulmonary emphysema is difficult and early direction is important. Multi-slice CT technology has been improving 3-D image analysis with higher body axis resolution and shorter scan time. And low-dose high accuracy scanning becomes available. Multi-slice CT image helps physicians with accurate measuring but huge volume of the image data takes time and cost. This paper is intended for computer added emphysema region analysis and proves effectiveness of proposed algorithm.

In order to satisfy the high resolution (3 to 10 cycles/mm) imaging requirements in neurovascular image-guided interventional (IGI) procedures, a micro-angiographic fluoroscope (MAF) is being developed to enable both rapid sequence angiography (15 fps) at high exposure levels (hundreds of μR/frame) as well as fluoroscopy at high frame rates (30 fps) and low exposure levels (5 to 20 μR/frame). The prototype MAF consists of a 350-μm-thick CsI(Tl) scintillator coupled by a 2:1 fiber-optical taper to an 18 mm diameter variable-gain light image intensifier with two-stage microchannel plate (MCP) viewed by a 12-bit, 1024x1024, 30 fps CCD camera with digital interface board. The optical set-up enables variation of effective pixel-size from 31 to 50 micron. The first frame lag of the MAF in fluoroscopic 30 fps mode (2:1 binning) was less than 0.8% at exposures of 5-23 μR/frame. MTF, NPS, and DQE in angiographic mode were measured for IEC standard spectrum RQA 5. At spatial frequencies of 4 and 10 cycles/mm the MTF was 14% and 1.5%, and the DQE was 12% and 1.2%, respectively, while the DQE(0) was 60%. Acquisition software was developed to acquire 15 fps angiography and 30 fps fluoroscopy for real-time dark field and flat field correction or real-time roadmapping. Images obtained with the MAF in small animal IGI procedures are demonstrated. The linearity versus x-ray intensity and MCP working range effects has been studied. We plan to expand the current 3.6 cm diameter field of view to 6 cm in the next model of the MAF.

The purpose of this paper is to provide a performance characterization of a new large field-of-view (LFOV) flat panel detector with a novel pixel design that has been optimized for both screening mammography and low dose advanced applications such as tomosynthesis. The measurements reported here were performed on prototype x-ray imagers for GE's upcoming LFOV mammography system. In addition to a light sensitive photodiode and a field effect transistor (FET), a storage capacitor has been added to each pixel in order to increase the dynamic range. In order to characterize the performance of the detector, measurements of the MTF, noise power spectrum, DQE, electronic noise, conversion factor, and lag were made. The results show that the new detector can deliver a DQE at 0 and 5 lp/mm of 72% and 28% while maintaining an MTF at 5 lp/mm of 30%. The addition of a storage capacitor at each pixel allows the conversion factor to be increased to reduce the noise floor - leading to a 400% extension of the dynamic range. Finally, a re-design of the FET and photodiode to reduce the time constants allows a 10X reduction in the lag that enables up to 4 frame per second imaging with less than 1% lag. This work represents the first results from a next generation large field of view a Si/CsI based x-ray imager for mammography and shows that a single detector can achieve high performance standards for both high dose screening and low dose, fast acquisition tomosynthesis simultaneously.

The contrast of X-ray imaging depends on the radiation energy and acquires its maximum value at a certain optimum energy typical for the object under investigation. Usually, higher energies result in reduced contrast, lower energies are absorbed in the object thus having a smaller probability of reaching the detector. Therefore, broad X-ray spectra contain non-optimal quanta to a large extent and deliver images with deteriorated contrast. Since investigations with monochromatic X-rays using synchrotrons are too complex and expensive for routine diagnostic imaging procedures, we propose a simpler approach. A conventional mammography system (Siemens Mammomat 300) with an X-ray tube with a molybdenum anode was supplemented with an X-ray HOPG monochromator (HOPG = Highly Oriented Pyrolytic Graphite) and an exit slit selecting those rays fulfilling Bragg’s condition. The detector is a CCD (Thales TH9570), 4092 x 200 pixels, 54 μm in size. At this slot-scan setup¹, measurements have been carried out at 17.5 keV as well as with a polychromatic spectrum with 35 kV tube voltage. The modulation transfer function (MTF) and the detective quantum efficiency (DQE) have been determined from images of a lead bar pattern and flat-field images. Both MTF and DQE depend on orientation (scan or detector direction) for the 17.5 keV monochromatic case. Above 3 mm^-1 the DQE values are smaller than those for polychromatic radiation. The contrast yielded by foils of different materials (Al, Cu, Y, Ag) has been studied. In all cases the monochromatic X-rays give rise to about twice the contrast of a polychromatic spectrum.

In this study, the beam stop technique was applied to obtain the scatter fraction values for an anthropomorphic breast phantom on a flat-panel full field mammography system. The phantom was equivalent to a compressed breast of 5 cm thickness with 50% glandular tissue content. The images were acquired at 28kVp with Mo/Mo target/filter combination and multiple mAs values with or without an anti-scatter grid. The one-dimensional and two-dimensional scatter fraction profiles of the breast phantom without a grid were plotted. The effect of mAs on scatter fraction calculation was investigated. It was found that in order to get a reliable measurement of scatter fraction, the mAs had to be above about 40mAs without a grid and 160mAs with a grid. In addition, the SNR values at 28kVp and 80mAs (AEC level for the phantom using the same imaging technique on a screen/film system) with and without a grid were compared with each other. The SNR with a grid was slightly smaller than that without a grid. The SNR improvement factor (K_SNR ) defined as the ratio of SNR without the grid to SNR with the grid was 0.976. The grid had the primary and scatter transmissions of 73% and 13% respectively.

The SenoScan full-field digital mammography scanner uses a scanning slot detector that is 10 mm wide and 220 mm long. The X-ray beam is collimated to just outside the area of the detector. One important advantage of slot scanning is its inherent scatter rejection. As previously reported, the SenoScan slot scatter rejection is better than that obtained using a 3.5:1 mammography grid, and somewhat worse than that with a 5:1 grid. Additional scatter reduction can potentially improve the contrast in images of thick breasts. We evaluate a custom-designed grid for the slot scanning system. The grid is one-dimensional, offering scatter rejection along the longitudinal axis of the detector. We evaluate the reduction in scatter fraction, grid absorption and changes in the signal-difference-to-noise ratio (SDNR). Based on phantom studies, our results show effective scatter reduction by the grid with minimal reduction of SDNR. Grid absorption and scatter elimination do not necessarily lead to an increase in patient dose, especially if there is a improvement in the number of digital values in the image that are within the useful dynamic range of the detector. A benefit of removing the scatter contribution is an improvement in system dynamic range, because electronic detector gain adjustments can compensate for the drop in the digital pixel values.

We are constructing and investigating a Scan Equalization Digital Radiography (SEDR) system using an a-Si:H based flat-panel detector. With this system, slot-scan imaging with regionally adjustable beam width is used to achieve scatter rejection and exposure equalization. As part of the SEDR system, we have developed and implemented an electronic aft-collimation technique, referred to as the Alternate Line Erasure and Readout (ALER), by modifying the electronics of an a-Si:H/a-Se based flat-panel detector to alter the sequence of image readout. Instead of reading the image line by line, the leading edge line of the scanning fan beam is reset to erase the scatter component accumulated prior to the arrival of the fan beam while the trailing edge line is read out to acquire the exposure signals integrated following the exposure of the scanning fan beam. This resetting and readout cycle is repeated and synchronized to the motion of the scanning fan beam. To guide the selection of the slot width in implementing the SEDR system, measurements of the scatter-to-primary ratio (SPR) and relative contrast-to-noise ratio (RCNR) were made and compared for various slot widths. Our preliminary testing has demonstrated that it is feasible to implement an electronic aft-collimation technique to effectively reject scattered radiation without attenuating the primary x-rays and without using a bulky, heavy aft-collimator. The SPR and RCNR measurements indicated that the performance of the slot-scan imaging technique improves with narrower slot width and is generally better than the anti-scatter grid method.

The purpose of this study is to compare the detection performance of three different mammography systems: screen/film (SF) combination, a-Si/CsI flat-panel (FP-), and charge-coupled device (CCD-) based systems. A 5-cm thick 50% adipose/50% glandular breast tissue equivalent slab phantom was used to provide an uniform background. Calcium carbonate grains of three different size groups were used to simulate microcalcifications (MCs): 112-125, 125-140, and 140-150 μm overlapping with the uniform background. Calcification images were acquired with the three mammography systems. Digital images were printed on hardcopy films. All film images were displayed on a mammographic viewer and reviewed by 5 mammographers. The visibility of the MC was rated with a 5-point confidence rating scale for each detection task, including the negative controls. Scores were averaged over all readers for various detectors and size groups. Receiver operating characteristic (ROC) analysis was performed and the areas under the ROC curves (Az’s) were computed for various imaging conditions. The results shows that (1) the FP-based system performed significantly better than the SF and CCD-based systems for individual size groups using ROC analysis (2) the FP-based system also performed significantly better than the SF and CCD-based systems for individual size groups using averaged confidence scale, and (3) the results obtained from the Az’s were largely correlated with these from confidence level scores. However, the correlation varied slightly among different imaging conditions.

Digital radiographs are often processed prior to printing or display. The algorithm used is generally a combination of contrast and edge enhancement applied in a spatially varying way. However, such enhancement often resulted in objectionable noise level in heavily attenuating regions, which may compromise the low contrast performance. We are developing and investigating a Scan Equalization Digital Radiography (SEDR) technique with which the transmitted x-ray exposure falling on the detector would be equalized and the image SNRs would be made more uniform. In this study, we simulated exposure equalization by acquiring a series of digital radiographs with incrementing mAs’ (0.25, 0.5, 1, 2, 4, 8) and then adding them with binary weighting factors to achieve equalized exposures over various regions of the image. The exposure-equalized image was then processed with two algorithms: local window/level optimization or edge enhancement using unsharp masking. The processed images with and without exposure equalization were then examined and compared with each other. For quantitative comparison, identically acquired images were used to obtain two sets of equalized and processed images. These two sets of images were subtracted from each other to generate a map of normalized noise for comparison. It is found that exposure equalization resulted in more uniform SNRs in both raw and processed images. Thus the noise levels in heavily attenuated regions were kept low and had less objectionable appearance for visualization of low contrast objects.

Clinical trials performed for the FDA’s Section 510k compliance submission of the Statscan digital, full-body, linear slit scanning diagnostic radiography system revealed that comparable diagnostic results with a commercial full-field screen film device were obtained with the Statscan using much lower radiation doses. For certain imaging procedures the doses for Statscan were as much as twenty to thirty times lower. However the results varied by a large amount and in particular the results for chest radiographs were anomalous in that the Statscan dose was less reduced. Whilst it is well known that slit scanning radiography has considerably lower radiation exposure than full-field devices due to its much lower scatter to primary ratio and also that digital radiography has the potential for lower radiation dosages, it was thought that that this alone did not fully account for the dose differences. This paper suggests that these dose differences, including the anomaly mentioned above, can be explained by considering the unique way that slit scanning is undertaken by Statscan i.e. by scanning the tube, detector, slit and collimators together along a linear path. The effect on measured skin entrance doses is explained and the dosage differences as affected by digital technology, higher DQE, slit scanning (low scatter to primary ratio) and linear slit scanning methods are quantified. Furthermore it is explained how the Statscan geometry leads an improved “skin sparing” effect.

Key features of a novel pin photodiode array structure built on 30-μm, 75-μm, and 100-μm thick single Silicon dies are discussed for the first time. Results of comparative studies of opto-electrical properties for a wide range of element sizes from ~ 200 μm square to ~ 1 mm square with the gaps between the adjacent elements as small as <20 μm are presented. The internal quantum efficiency was close to 100%, crosstalk was smaller than 0.01% within the spectral range 400 to 800 nm. The crosstalk remained lower than 0.1% even in the case when the illuminated element was electrically isolated (open contact). Furthermore very low leakage current and high shunt resistance (above 1 GΩ) are characteristic for these devices. The results are imperative for creating of high quality imaging systems.

The effect of finite open-loop gain in charge amplifiers in digital radiography (DR) is analyzed. Practical charge amplifiers are usually integrated into silicon chips and commonly cater to 128 columns. High gains in charge amplifiers have a cost that is associated with greater chip count, greater power dissipation, and sometimes, reduced bandwidth. Even an open-loop gain of 1000 in a charge amplifier can lead to visible artifacts in a DR system, for temperature drifts of a fraction of a degree between calibration of a panel and subsequent usage, while acquiring a radiographic image. Furthermore, small gains in a charge amplifier can rob charge from the pixel capacitor and create residual charge, predominantly in the column capacitances. This can reduce the effective signal-to-noise ratios (SNRs). Analytical models are developed to illustrate the effects of the finite gain in charge amplifiers. Methods are suggested to decrease the deleterious effects of finite gains in charge amplifiers.

One of the results of the latest developments in x-ray tube and detector technology, is the enabling of computed tomography (CT) as a strong non-invasive imaging modality for a new set of clinical applications including cardiac and brain imaging. A common theme among the applications is an ability to have wide anatomical coverage in a single rotation. Large coverage in CT is expected to bring significant diagnostic value in clinical field, especially in cardiac, trauma, pediatric, neuro, angiography, Stroke WorkUp and pulmonary applications. This demand, in turn, creates a need for tile-able and scalable detector design. In this paper, we introduce the design of a new diode, a crucial part of the detector, discuss how it enables wide coverage, its performance in terms of cross-talk, light output response, maximized geometric efficiency, and other CT requirements, and compare it to the traditional design which is front-illuminated diode. We ran extensive simulation and measurement experiments to study the geometric efficiency and assess the cross talk and all other performance parameters Critical To Quality (CTQs) with both designs. We modeled x-ray scattering in the scintillator, light scattering through the septa and optical coupler, and electrical cross talk. We tested the design with phantoms and clinical experiments on a scanner (LightSpeed VCT, GE Healthcare Technologies, Waukesha, WI, USA). Our preliminary results indicate that the new diode design performs as well as the traditional in terms of cross talk and other CTQs. It, also, yields better geometric efficiency and enables tile-able detector design, which is crucial for the VCT. We introduced a new diode design, which is an essential enabler for VCT. We demonstrated the new design is superior to the traditional design for the clinically relevant performance measures.

In rotational tomotherapy a high energy (6 MeV) photon beam irradiates the patient. A CT detector placed behind the patient is used to establish the position of the patient and the dose delivered. A possible detector design uses amorphous selenium (a-Se) as the x-ray to charge conversion medium requiring a detailed investigation of the change in x-ray sensitivity due to exposure to radiation in a-Se. Our novel experimental method called x-ray time of flight provides instantaneous measurements of x-ray sensitivity and charge transport parameters in a-Se films. The method analyzes the current from a-Se samples in response to single short pulses of LINAC radiation. X-ray sensitivity was observed to decrease substantially with large exposures (e.g. more than 50% after 4 Gy accumulated over ~5min) and to completely recover over <48h. The mechanisms responsible were studied from the kinetics of the measured current waveforms. On the basis of experimental data, a model for dose dependent sensitivity was formulated taking into account carrier trapping, re-distribution of electric field in the a-Se film due to space charge and evolving recombination processes. In principle quantitative comparison of experimental and theoretical characteristics will permit the determination of the main material parameters (carrier mobility, deep trapping lifetime), as well as the generation rate of carriers by x-rays. Thus a basis for the development of a practical a-Se based megavoltage CT detector has been investigated.

This report presents a system model of the STATSCAN slit scanning full body radiography machine. A Cascaded Linear Systems model of the detector was developed and the theoretical DQE, MTF and NPS were compared to measured values for the RQA9 beam quality described in IEC 62220-1. The effect on detector DQE of various system parameters such as coupling efficiency, CCD noise and pixel binning was quantified. System performance for various thicknesses of Gd₂O₂S:Tb was analyzed. The notion of a “System DQE” has been suggested by several authors to facilitate the comparison of overall systems. An expression for the overall “System DQE” was developed by including the effects of scattered radiation, grid attenuation and focal spot unsharpness in the cascaded model. Scattered radiation was quantified as a function of system geometry parameters and was treated as an “additive noise stage”. A realistic model of the focal spot was used to calculate the MTF due to beam divergence in the scan direction and focal spot unsharpness in the slit direction. It was found that the “System DQE” is a valuable parameter for the purpose of comparing gridless slit scanning system performance to conventional geometry system performance.

The presampling modulation transfer function (MTF) of a digital imaging system is commonly determined by measuring the system’s line spread function (LSF) using a narrow slit or differentiating the detector’s edge spread function (ESF) with an edge device. The slit method requires precise fabrication and alignment of a slit as well as a high radiation exposure. The edge method [3] is a complicated image processing procedure, requiring determination of the edge angle, reprojection, sub-binning, smoothing and differentiating the ESF, and spectral estimation. In this paper, a simple method is employed to evaluate the MTF using an edge device. The image processing procedures required by this method involve simply the determination of the over-sampling rate and the Fourier transform of the modified ESF. Differentiation and signal to noise ratio (SNR) improvement are jointly applied in the Fourier domain. The MTFs obtained by this simple method are compared to the theoretical MTF and the previously proposed more complicated edge method. The experimental results show that the proposed method provides a simple, accurate and convenient measurement of the presampling MTF for digital imaging systems.

Calculation of the modulation transfer function (MTF) is a multi-step procedure. At each step in the calculation, the algorithms can have intrinsic errors which are independent of the imaging system or physics. We designed a software tool with a graphical user interface to facilitate calculation of MTF and the analysis of accuracy in those calculations. To minimize the source of errors, simulated edge images without any noise or artifacts were used. We first examined the accuracy of a commonly used edge-slope estimation algorithm; namely line-by-line differentiation followed by a linear regression fit. The influence of edge length and edge phase on the linear regression algorithm is demonstrated. Furthermore, the relationship of edge-slope estimation error and MTF error are illustrated. We compared the performance of two kernels, [-1,1] and [-1,0,1], in the computation of the line spread function (LSF) from finite element differentiation of the edge spread function (ESF). We found that there is no practical advantage in choosing the [-1,0,1] kernel, as recommended by IEC. However, a correction for finite element differentiation should be applied; otherwise, there is a measurable error in the MTF. Finally, we added noise into the edge images and compared the performance of two noise reduction methods on the ESF; convolution with a boxcar kernel and a monotonicity constraint. The former method always produces MTF error higher than 4% up to the sampling frequency, while the latter was consistently less than 1%.

The determination of point spread function ( PSF ) is essential for the assessment of the performance of multi-slice CT corresponding to particular scanning or reconstruction condition. We studied the accuracy of the method for determination of PSF. The precision were evaluated with use of special made phantom and computer simulations. The PSF of some kinds of multi-slice CT were evaluated under the several scanning or reconstruction conditions. The performances were compared with each other and discussed. The PSF ( x, y, z ) was measured with separate components of PSF ( x, y ) and LSF (z). In the measurement of PSF ( x, y ) and LSF (z), we used Boone phantom and delta phantom respectively. The PSF ( x, y ) was calculated from LSF (x) and LSF (y). The LSF ( z ) obtained from deltal phantom was added for the calculation of PSF ( x, y, z ) with use of the following formula. PSF ( x, y, z ) = PSF ( x, y ) • LSF ( Z ) = LSF ( x ) • LSF ( y ) • LSF ( z ). The precision of the obtained LSF and PSF were evaluated by phantom experiment with computer simulation using following equations. I ( x ) = LSF ( x ) * O ( x ). [equation]. The precision of the measured PSF ( x, y ) was evaluated to be a good approximation for the image of the blood-vessel phantom. The results propose the method for the evaluation of the precision of measured point spread function. At the same time this results suggests the method of the calculation of the CT image affected by spatial resolution.

The international standard IEC 62220-1 about DQE measurement of digital X-ray equipment was published in 2003, but mammography systems aren’t applied to this IEC standard because the factor affect measurement is complicated. Especially, the influence to the pre-sampling MTF by edge method when X-ray beam is oblique to detector. The influence of nonuniformity of x-ray intensity by the heel effect on digital Wiener spectrum (WS) doesn’t become clear. A 0.1mm-thick tungsten edge was imaged in the position where X-ray beam was perpendicular to detector plane and in 6cm from chest wall, respectively. And the pre-sampling MTFs were obtained from these edge images. The calculation area of the digital WS within irradiation area was moved in parallel direction to X-ray tube axis, and the digital WS were calculated. The pre-sampling MTFs and the digital WS are calculated by the method based on the IEC proposal. We used MAMMOMAT3000(SIEMENS), MGU-100B(TOSHIBA), M-IV(LORAD) and Senographe DMR+(GE) as X-ray generator. Images were obtained by FCR PROFECT CS (Fujifilm medical). In all equipments and both arrangement directions of the edge test device, pre-sampling MTFs are almost the same, even if the arrangement places of the edge test device varied. In all equipments, when the calculation area was moved about 10cm, the digital WS of the anode side was higher 7.2-17.9% than those of the cathode side. It was found that the dose of anode side was lower about 20% than cathode side from the profile of an exposure image. We think that digital WS modified the nonuniformity of the dose by the heel effect is obtained by multiplying the digital WS by the compensation coefficient obtained by the dose profile, in low spatial frequency.

Recently, x-ray computed tomography (CT) systems were developed dramatically; e.g. a multi-detector-row CT or a 4-dimensional CT, but it has been expressed anxiety that a patient dose is increased. Analysis of x-ray spectrum is important for quality assurance and quality control of radiographic systems to estimate a quality of an imaging system and to decrease a patient dose. The aim of this study is to measure the x-ray spectra of CT system under clinical conditions using a high resolution Schottky CdTe detector. When measuring diagnostic x-ray spectra, the long distance from the x-ray source to the detector is requested for reducing a number of photons detected per unit time to prevent pile-up of the detector. However, that is very difficult to set up the long source-to-detector distance in a gantry of a CT unit. For resolving this problem, the Compton spectroscopy is very suitable. Using this method, a number of photons detected per unit time can be reduced by detecting the scattered x-ray photons. If the 90° scattered photons can be detected, the energy correction and reconstruction of spectra can be calculated easily by use of the Klein-Nishina formula. So we attempt to acquire the primary x-ray spectra in the gantry of the CT unit by using Compton spectroscopy under a clinical (tube-rotating) condition. Moreover, to estimate the variation of x-ray spectra owe to changing position in the gantry, we measured the x-ray spectra and exposure doses at various points in the gantry.

Acquisition data and treatments for quality controls of gamma cameras and Positron Emission Tomography (PET) cameras are commonly performed with dedicated program packages, which are running only on manufactured computers and differ from each other, depending on camera company and program versions. The aim of this work was to develop a free open-source program (written in JAVA language) to analyze data for quality control of gamma cameras and PET cameras. The program is based on the free application software ImageJ and can be easily loaded on any computer operating system (OS) and thus on any type of computer in every nuclear medicine department. Based on standard parameters of quality control, this program includes 1) for gamma camera: a rotation center control (extracted from the American Association of Physics in Medicine, AAPM, norms) and two uniformity controls (extracted from the Institute of Physics and Engineering in Medicine, IPEM, and National Electronic Manufacturers Association, NEMA, norms). 2) For PET systems, three quality controls recently defined by the French Medical Physicist Society (SFPM), i.e. spatial resolution and uniformity in a reconstructed slice and scatter fraction, are included. The determination of spatial resolution (thanks to the Point Spread Function, PSF, acquisition) allows to compute the Modulation Transfer Function (MTF) in both modalities of cameras. All the control functions are included in a tool box which is a free ImageJ plugin and could be soon downloaded from Internet. Besides, this program offers the possibility to save on HTML format the uniformity quality control results and a warning can be set to automatically inform users in case of abnormal results. The architecture of the program allows users to easily add any other specific quality control program. Finally, this toolkit is an easy and robust tool to perform quality control on gamma cameras and PET cameras based on standard computation parameters, is free, run on any type of computer and will soon be downloadable from the net (http://rsb.info.nih.gov/ij/plugins or http://nucleartoolkit.free.fr).

The excised rat kidney slices were investigated using the diffraction enhanced imaging (DEI) method under the different FWHM of the rocking curve with 33.7keV x-ray, for the first time. The narrower FWHM of the rocking curve is used in the DEI imaging, the finer structure of the sample can be clearly distinguished and more details can be seen in the DEI image. Tuning the rocking curve width between 1.7μrad and 0.31μrad was done with small loss of peak intensity using a Si (220) double-crystal analyzer. The reason related with the influence of the FWHM of the rocking curve to the contrast resolution of DEI method is discussed. For the thin sample, how small deflected angle can be distinguished determines how small difference of density can be distinguished in the DEI imaging.

Diffraction Enhanced Imaging (DEI) is an x-ray phase contrast technique, which is showing great promise for a number of medical imaging problems. For a source it relies on a highly collimated flux of monochromatic x-rays, which is currently only available at synchrotron radiation facilities. Phase shifts occurring as the wave passes through the object are made visible using Bragg diffraction from a post-sample analyser optic. In early 2004 the DEI system on the bending-magnet beam line 7.6 of the Daresbury SRS was used for the first time to image a number of small medical specimens. This paper will report on the performance of the system and the results of these initial studies. A new DEI instrument is currently in the design phase. This facility will be integrated on wiggler station 9.4 on the SRS allowing access to shorter x-ray wavelengths and greater flux. A progress report on the design features and implementation of this system will be given.

Intensity diffraction tomography (IDT)is a propagation-based phase-contrast tomography imaging method. In this work, we develop an IDT reconstruction theory for measurement geometries that employ tilted detector planes. The conventional IDT reconstruction theory is contained as a special case where the detector planes are perpendicular to the direction of the probing plane-waves. The resulting reconstruction algorithms are implemented numerically, and computer-simulation studies are conducted to demonstrate their validity and robustness to data noise.

Amorphous selenium (a-Se) flat-panel digital mammography detectors are being investigated for tomosynthesis, which poses tremendously challenges on the detector temporal and low dose performance. Our previous investigation has demonstrated that a-Se detectors provide adequate temporal performance (lag and ghosting) for tomosynthesis, however its detective quantum efficiency (DQE) at 1 mR (1/10 of average exposure in screening mammography) was only ~1/3 of the value at 10 mR due to electronic noise. Before engineering methods can be developed to overcome this problem, optimization of imaging parameters, such as x-ray spectrum and anti-scatter grid, can greatly improve the detector performance at the low dose used in tomosynthesis. The purpose of this paper is to determine the optimal x-ray spectrum and whether an anti-scatter grid is beneficial in tomosynthesis. The SNR of a 200 μm microcalcification within the breast was calculated as a function of x-ray spectra. Two target materials (Mo and W) were used. The density and thickness of the breast were varied. The scatter to primary ratio behind the breast with and without grid was calculated. The detector performance of a state-of-the-art a-Se digital mammography detector with 85 micron pixel size was incorporated in the calculation of SNR. The total breast dose was kept constant at 1.6 mGy. Our results showed that for tomosynthesis with 11 acquisition views, the optimal kVp is at least 2-3 kVp higher than the optimum for screening mammography. In the extreme case of an 8 cm dense breast, the optimal spectrum was 39 kVp (W/Rh), which was 9 kVp higher than the optimal kVp when detector noise is negligible. W/Rh was found to be the optimal target filter combination for all breast thicknesses (2-8 cm). Our results also showed that grid has no clear advantage even for breast thickness of 8 cm.

The purpose of this work is to report the performance of an amorphous selenium (a-Se) based flat-panel x-ray imager under development for application in digital breast tomosynthesis. This detector is designed to perform both in the conventional Full Field Digital Mammography (FFDM) mode and the tomosynthesis mode. The large area 24 x 29 cm detector achieves rapid image acquisition rates of up to 4 frames per second with minimal trapped charge induced effects such as ghost or lag images of previously acquired objects. In this work, a new a-Se/TFT detector layer structure is evaluated. The design uses a top conductive layer in direct contact with the a-Se x-ray detection layer. The simple structure has few layers and minimal hole and electron trapping effects. Prototype detectors were built to investigate the basic image performance of this new a-Se/TFT detector. Image signal generation, image ghosting, image lag, and detector DQE were studied. For digital mammography applications, the residual image ghosting was less than 1% at 30 seconds elapsed time. DQE, measured at a field of 5.15 V/um, showed significantly higher values over previously reported data, especially at low exposure levels. For digital breast tomosynthesis, the image lag at dynamic readout rate was < 0.6 % at 0.5-second elapsed time. A prototype tomosynthesis system is being developed utilizing this new a-Se/TFT detector.

A method for calibrating camera geometry is described. This method has been used to implement synthetic tomography on a commercially available full-field digital mammography system. The method utilizes a phantom containing six point-like calibration objects whose positions are approximately known. The image of five calibration objects in a given projection allows an associated projection matrix to be determined up to one free parameter. By using the positions of the shadows of the sixth calibration object in three or more views, one can fit the remaining free parameter associated with each view and the position of the sixth calibration object relative to the first five. Uncertainty in the position or geometry of the phantom does not affect the geometric consistency, thus tomograms produced by back-projection suffer no blurring from errors in the determination of camera geometry. Uncertainties in the position or geometry of the phantom result in proportionate translations or distortions of the tomograms. For a tomogram corresponding to a plane containing an object, the positions of the backprojections of the shadows of the object are consistent to the same precision as the measurements of the shadows in each projection, i.e., the positions of the backprojections differ by about the size of the pixel spacing in the detector.

We present preliminary investigations that examine the feasibility of incorporating digital tomosynthesis into radiation oncology practice with the use of kilovoltage on-board imagers (OBI). Modern radiation oncology linear accelerators now include hardware options for the addition of OBI for on-line patient setup verification. These systems include an x-ray tube and detector mounted directly on the accelerator gantry that rotate with the same isocenter. Applications include cone beam computed tomography (CBCT), fluoroscopy, and radiographs to examine daily patient positioning to determine if the patient is in the same location as the treatment plan. While CBCT provides the greatest anatomical detail, this approach is limited by long acquisition and reconstruction times and higher patient dose. We propose to examine the use of tomosynthesis reconstructed volumetric data from limited angle projection images for short imaging time and reduced patient dose. Initial data uses 61 projection images acquired over an isocentric arc of twenty degrees with the detector approximately fifty-four centimeters from isocenter. A modified filtered back projection technique, which included a mathematical correction for isocentric motion, was used to reconstruct volume images. These images will be visually and mathematically compared to volumetric computed tomography images to determine efficacy of this system for daily patient positioning verification. Initial images using the tomosynthesis reconstruction technique show much promise and bode well for effective daily patient positioning verification with reduced patient dose and imaging time. Additionally, the fast image acquisition may allow for a single breath hold imaging sequence, which will have no breath motion.

The purpose of this paper is to investigate the use of electron-beam Computed Tomography (EBCT) dual energy scanning for improved differentiation of calcified coronary arteries from iodinated-contrasted blood, in fast moving cardiac vessels. The dual energy scanning technique can lead to an improved cardiac examination in a single breath hold with more robust calcium scoring and better vessel characterization. Dual energy can be used for material discrimination in CT imaging to differentiate materials with similar CT number, but different material attenuation properties. Mis-registration is the primary source of error in a dual energy application, since acquisitions have to be made at each energy, and motion between the acquisitions causes inconsistencies in the decomposition algorithm, which may lead to artifacts in the resultant images. Using EBCT to quickly switch x-ray source peak voltage potential (kVp), the mis-registration of patient anatomy is minimized since acquisitions at both energy spectra are completed in one study at the same cardiac phase. Two protocols for scanning the moving heart using EBCT were designed to minimize registration issues. Material basis function decomposition was used to differentiate regions containing calcium and iodine in the image. We find that this protocol is superior to CT imaging at one energy spectrum in discriminating calcium from contrast-enhanced lumen. Using dual energy EBCT scanning can enable accurate calcium scoring, and angiography applications to be performed in one exam.

The implementation of contrast-enhanced dual-energy digital subtraction mammography may lead to better identification of breast tumors, and thus provide a lower cost and more widely available alternative to breast MRI. This technique involves the acquisition of low- and high-energy images after the IV administration of iodinated contrast agent. In this study, the effect of the beam energy (kVp) was examined using the CNR²/dose metric, where CNR is the contrast-to-noise ratio and dose implies the mean glandular dose. The mean glandular dose was calculated using parameterized normalized glandular dose coefficients (DgN), which allowed the computation of the mean glandular dose for the modeled spectra considered in this study, coupled with incident kerma measurements. Optimization studies were performed using a dedicated cone-beam breast CT scanner designed and fabricated in our laboratory, with the system operating in stationary imaging mode. A flat tissue-equivalent phantom (7.5 cm in thickness) was placed at the isocenter of the scanner, and an air gap of 34.5 cm was used in lieu of a grid. Dilute iodine-based contrast agent was introduced into the phantoms using plastic vials. Data were acquired from 40 to 90 kVp at 10 kVp intervals. Due to the low mA available on the breast CT system, a large number of images (1000) were acquired in fluoroscopic mode, which allowed us to match the dose and noise properties for each kVp combinations by changing the number of images used for averaging. Preliminary results demonstrate that the best CNR2/dose is achieved with a 50 kVp low-energy image and a 90 kVp high-energy image. Consequently, radiation doses for contrast-enhanced mammography should be far lower than regular mammography. Since the spatial resolution requirements should also be lower than regular mammography, dual-energy contrast-enhanced mammography, when performed using the optimal technique factor, may indeed provide very similar diagnostic information as breast MRI but at significantly reduced costs.

The image contrast among tumor tissue and non-tumor tissue in a breast volume reconstructed by cone-beam computed tomography (CBCT) breast imaging is usually very small. Striving for contrast enhancement by exploiting the energy dependent x-ray attenuation, we herein report a dual-kVp CBCT breast imaging modality. Based on dual-kVp conebeam scan and dual-basis-material decomposition, we represent breast tissues in a two-feature space that is spanned by two basis material functions. Though linear independent, the basis material functions are partially correlated or nonorthogonal. Therefore, the dual-material-equivalent decomposition is essentially a sort of non-orthogonal expansion, which is more useful for material classification than for quantitative measurement. On the other hand, the curse of dimensionality discourages high-dimensional feature space that may be spanned by using more basis material functions. For optimizing the material space, we suggest the use of two material spaces (different in basis functions). The first material space is spanned by {bone, polyethene}, which covers a wide range of x-ray attenuation, from calcification to soft tissue. The second material space is spanned by {teflon, fat}, which covers a small range, suitable for low-contrast soft tissue discrimination. Based on x-ray energy-dependent attenuation of materials and x-ray spectra, we simulated the dual-kVp CBCT breast imaging modality with pre-reconstruction scheme, thereby showing the feasibility for image contrast enhancement by dual-kVp technique. With the test materials: breast tissue, water, and soft tissue, the simulation showed that the dual-kVp technique could double the image contrast in this particular case. Through experiment with our CBCT prototype, we demonstrated dual-kVp volume subtraction, showing the energy-dependent x-ray attenuation.

The LEXXOS (DMS, Montpellier, France) is the first axial and total body cone beam bone densitometer using a 2D digital radiographic detector. Technical principles and performances for BMD measurements have been presented in previous papers. Bone densitometers are also used on small animals for drug development. In this paper, we show how the LEXXOS system can be adapted to small animals examinations, and its performances are evaluated. At first, in order to take advantage of the whole area of the digital flat panel X-ray detector, the geometrical configuration has been adapted. Secondly, as small animals present low BMD, a specific dual energy calibration has been defined. This adapted system has then been evaluated on two sets of mice: six reference mice and six ovariectomized mice. Each month, these two populations have been examined and the total body BMD has been measured. This evaluation has shown that the right order of BMD magnitude has been obtained and, as expected, BMD increases on the two sets until age of puberty and after this period, decreases significantly for the ovariectomized set. Moreover, the bone image obtained by dual energy processing on LEXXOS presents a radiographic image quality providing with useful complementary information on bone morphometry and architecture.

Overlapping fibroglandular tissue structures may obscure small calcifications, essential to the early detection of breast cancer. Dual-energy digital mammography (DEDM), where separate low- and high-energy images are acquired and synthesized to cancel the tissue structures, may improve the ability to detect and visualize calcifications amidst fibroglandular structures. We have developed and implemented a DEDM technique under full-field imaging conditions using a commercially available flat-panel based digital mammography system. We have developed techniques to suppress residual structures due to scatter contamination and non-uniformity in the x-ray field and detector response in our DEDM implementation. The total mean-glandular dose from the low- and high-energy images was constrained to be similar to screening examination levels. The low- and high-energy images were combined using a calibrated nonlinear (cubic) mapping function to generate the calcification images. To evaluate the dual-energy calcification images, we have designed a special phantom with calcium carbonate crystals to simulate calcifications of different sizes superimposed with a 5 cm thick breast-tissue-equivalent material with a continuously varying glandular-tissue ratio from 0.0 to 1.0. The suppression of tissue-structure background by dual-energy imaging comes with the cost of increased noise in the dual-energy images. We report on the effects of different image processing techniques on the dual-energy image signal and noise levels. The effects of image processing on the calcification contrast-to-noise ratios are also presented.

X-ray storage phosphors have several advantages over traditional films as well as digital X-ray detectors based on thin-film transistors (TFT). Commercially used storage phosphors do not have high resolution due to light scattering from powder grains. To solve this problem, we have developed storage phosphor plates based on modified fluorozirconate (ZBLAN) glasses. The newly developed imaging plates are “grainless” and, therefore, can significantly reduce light scattering and improve image resolution. To study the structure and image performance of the novel storage phosphor plates, we conducted X-ray diffraction (XRD) and X-ray imaging analyses at the Advanced Photon Source, Argonne National Laboratory. The XRD results show that BaCl₂ crystallites are embedded in the glass matrix. These crystallites enlarge and are under residual stress after heat treatment. The X-ray imaging study shows that these storage phosphor plates have a much better resolution than a commercially used storage phosphor screen. The results also show that some of the glass ceramics are high-resolution scintillators. Our study demonstrates that these fluorozirconate-based glass ceramics are a promising candidate for high-resolution digital X-ray detectors for both medical and scientific research purposes.