Early detection of breast cancer largely contributes to an improved prognosis and results in reduced disease mortality. The breast cancer screening technique used today in clinics is known as dual-view digital mammography. Its limitation is that it only provides two images of the breast, which can explain why up to 20% of breast tumors are not detectable on mammograms. Mammograms can also appear abnormal although no breast cancers are actually present. Computed tomography (CT) allows a precise 3D visualization of the human body, but it cannot be routinely used for breast cancer diagnosis because of the breast's high radiosensitivity. This vulnerability significantly reduces the benefit-risk ratio owing to the potential for tumor development resulting from the exam itself.
Recognizing these limitations, we developed a way to produce 3D x-ray images of the breast at a radiation dose that is lower than 2D radiographies.1 We performed phase-contrast x-ray tomography at 60keV of a whole human breast at the biomedical beamline of the European Synchrotron Radiation Facility (ESRF) in France. We reconstructed images by applying the novel mathematical equally sloped tomography (EST) algorithm using 512 projections to produce 3D images of the tissue at a resolution higher than that used in clinical CT (pixel size of 92μm versus more than 200μm for clinical CT scanners) and at a dose lower than that of dual view mammography (which is typically about 3mGy). The new method allowed a clear and precise identification of a malignant cancer. According to a blind evaluation by five experienced radiologists, this method can reduce the radiation dose and acquisition time by ∼74% relative to conventional phase-contrast x-ray tomography while maintaining high resolution and contrast.
Despite the significant reduction in the number of projections used (corresponding to important savings in dose and acquisition time), radiologists ranked the generated images as having the highest sharpness, contrast, and overall image quality compared to 3D images of breast tissue created through other standard methods based on the filtered back projection (FBP) algorithm (see Figure 1).
Figure 1. Comparison between a conventional computed tomography (CT) scan of the breast based on the filtered back projection algorithm and one based on the equally sloped tomography algorithm with phase-contrast imaging. In the second image, the tumor is highlighted in red. The radiation dose needed for the scans is shown at the bottom of each image.
Improved detection of breast cancer using CT scans now seems possible thanks to the combination of three technologies: high energy x-rays, phase-contrast imaging, and the use of the sophisticated EST algorithm to reconstruct the CT images from x-ray data. High-energy x-rays render tissues more transparent, which reduces the radiation dose that is deposited by a factor of six. Phase-contrast imaging, which our teams at the ESRF and the Ludwig Maximilians University are working on, may allow contrast from the x-ray's phase modulation (or refraction) even if amplitude (i.e., absorption) modulation is weak or absent. This is possible with a radiation dose similar to, or even smaller than, that used in conventional absorption radiography.
The final element is EST, originally developed by researchers at the University of California-Los Angeles, is a computational method based on successive approximations that iterate back and forth between real and Fourier space using the pseudo-polar fast Fourier transform. At each iteration, physical constraints, including the sample boundary and the non-negativity of the sample structure, are enforced in real space, while the Fourier transform of the measured projections are imposed in Fourier space. As a result, EST needs four times less radiation than the FBP to produce images of the same quality. An important step toward the clinical implementation of this method is the requirement that compact x-ray sources be able to obtain quasi-monochromatic x-rays with flux densities between those obtained at large-scale synchrotron radiation facilities and at clinical x-ray generators. Fortunately, such compact x-ray systems are currently under rapid development worldwide: through compact synchrotron radiation, tabletop high harmonic generation, and Compton backscattering. Finally, although we used a human breast cancer sample as proof of principle in this study, this method can be applied to other medical tomography fields where high resolution, high contrast, low radiation, and fast data acquisition are crucially needed.
Figure 2. Image quality comparison of the phase-contrast computed tomography images of the whole breast (9cm diameter) reconstructed using the conventional filtered back projection (FBP) algorithm and the equally sloped tomograpy (EST) algorithm. (A) A 92μm-thick sagittal slice of the FBP reconstruction using 2000 projections, (B) the EST using 512 projections, and (C) EST 200. The yellow rectangle indicates a tumor region. FBP 2000 and EST 512 have the highest image quality. The EST 200 (with one-tenth of the dose necessary to reconstruct the FBP2000) still shows a high contrast.
In conclusion, we have demonstrated that, compared with current clinical mammography, the EST-based phase-contrast tomography method can not only provide 3D data on soft tissue and tumors at a spatial resolution 2–3 times higher than present hospital scanners, but also deliver a radiation dose that is about 25 times lower. This new technique can become a powerful tool for diagnosing breast cancer and allow clinicians to battle the disease more effectively.
The future goals of our group include the application of the phase-contrast imaging technique combined with the EST algorithm to other organs to contribute to improved high-resolution and low-dose diagnosis in a broad variety of medical cases.
We thank the European Synchrotron Radiation Facility for providing the experimental facilities. Our work was partially supported by the University of California's Discovery/TomoSoft Technologies Grant IT107-10166, National Institutes of Health Grant GM081409-01A1, and the Deutsche Forschungsgemeinschaft—Cluster of Excellence Munich—Centre for Advanced Photonics EXE158.
Paola Coan, Susanne Grandl, Anikó Sztrókay-Gaul
Ludwig Maximilians University
Paola Coan is an associate professor leading a research group using intense and collimated x-ray sources for medical diagnostics at the Ludwig Maximilians University in Germany. She received her PhD in physics at the J. Fourier University and the European Synchrotron Radiation Facility in France with a thesis on phase-contrast imaging techniques for biomedical applications.
Alberto Bravin, Emmanuel Brun
European Synchrotron Radiation Facility
Alberto Bravin is chief physicist of the biomedical beamline at the European Synchrotron Radiation Facility. He is leading the preclinical phase-contrast imaging programs applied to mammography and cartilage studies. He co-authored more than 120 peer-reviewed scientific papers on medical applications of synchrotron radiation.
Jianwei Miao, Yunzhe Zhao
University of California-Los Angeles
Los Angeles, United States
Jianwei Miao is an internationally renowned pioneer in developing novel imaging methods with x-rays and electrons. In 1999, he performed the first experiment on extending x-ray crystallography to allow structural determination of non-crystalline specimens. More recently, he developed an innovative electron tomography method for 3D imaging of local structures at atomic resolution.
1. Y. Zhao, E. Brun, P. Coan, Z. Huang, A. Sztrókay, P. C. Diemoz, S. Liebhardt, A. Mittone, S. Gasilov, J. Miao, A. Bravin, High-resolution, low-dose phase contrast X-ray tomography for 3D diagnosis of human breast cancers,
109(45), p. 18290-18294, 2012.