SPIE Startup Challenge 2015 Founding Partner - JENOPTIK Get updates from SPIE Newsroom
  • Newsroom Home
  • Astronomy
  • Biomedical Optics & Medical Imaging
  • Defense & Security
  • Electronic Imaging & Signal Processing
  • Illumination & Displays
  • Lasers & Sources
  • Micro/Nano Lithography
  • Nanotechnology
  • Optical Design & Engineering
  • Optoelectronics & Communications
  • Remote Sensing
  • Sensing & Measurement
  • Solar & Alternative Energy
  • Sign up for Newsroom E-Alerts
  • Information for:
SPIE Photonics West 2017 | Register Today

OPIE 2017

OPIC 2017

SPIE Defense + Commercial Sensing 2017 | Register Today




Print PageEmail PageView PDF

Biomedical Optics & Medical Imaging

Design proposed for a combined MRI/computed-tomography scanner

Researchers may be able to better guide radiation therapy, analyze atherosclerotic plaques, and assess acute strokes and brain injuries using simultaneously captured multi-modal images.
11 June 2013, SPIE Newsroom. DOI: 10.1117/2.1201305.004860

Computed tomography (CT) and MRI are arguably the two most important imaging approaches in modern hospitals and clinics. CT offers fast scanning speed and high spatial resolution, but suffers from poor soft-tissue contrast and a substantial radiation dose. MRI, on the other hand, provides superior soft-tissue contrast, and functional and molecular imaging capabilities, but often suffers from long scan time and sub-optimal geometrical accuracy. MRI is also difficult to quantify in any absolute units, and has much greater sensitivity than specificity. In some clinical scenarios—such as guiding radiation therapy, analyzing atherosclerotic plaques, and assessing acute strokes and brain injuries—CT and magnetic resonance (MR) images can be synergistic when captured simultaneously. However, because clinicians typically collect CT and MR images independently at different times and retrospectively register them based on a number of approximated assumptions, residual errors can be introduced. These errors can originate from non-repeatable contrast dynamics, organ motion, patient posture, signal nonlinearity, and inconsistent contrast mechanisms between CT and MRI.

A combined CT-MRI scanner would reduce residual errors, enable simultaneous anatomical and functional imaging, and minimize exposure to ionizing radiation. Although multimodal imaging systems have become increasingly popular, a combined CT-MRI system has not yet been attempted. Two major obstacles in designing such a system are the bulkiness of the imaging scanners and the conflict in imaging physics, namely CT's rotating metallic parts and MRI's magnetic fields. Overcoming these challenges, we recently created a design for a combined CT-MRI scanner.1 With the dual-modality system, CT provides a snapshot of body's structures, while MRI reveals cellular and molecular features, blood flow, and soft-tissue. The hybrid scanner is designed to collect high-contrast images of many features with adequate spatial and temporal resolution and within the same coordinate system.

We designed the CT-MRI scanner based on realistic numerical simulation and engineering optimization. The key idea behind our system is that each imaging modality focuses on a relatively small, common region of interest (ROI). Because traditional CT methods cannot exactly reconstruct an interior ROI solely from truncated projections along x-rays through the ROI, current CT architectures have contained detectors that fully cover a transverse slice. A technique we developed called interior tomography2 enables theoretically exact reconstruction over an ROI from local data aided by practical prior knowledge. Applying interior tomography to MRI allows us to use a uniform magnetic field over a small imaging region, which is one way our design compensates for the incompatibility between the MRI's powerful magnets and the CT scanner's rotating metal parts. The compression makes the necessary room to place the tomographic modalities tightly together, and allows them to operate in parallel, achieving synchrony in space and time.

We have proposed three potential applications for our MRI-CT scanner: guiding radiation therapy, analyzing atherosclerotic plaques, and assessing acute strokes and brain injuries. Radiation therapy can be curative if the cancer is localized, and the therapy is precise. In 2010, Rasch et al. reported decreased observer variation of target delineation in nasopharynx cancer with retrospectively registered CT and MR images.3 In 2013, Shukla et al. compared image-guided radiation therapy techniques and assessed the time from imaging to registration, as well as residual errors, which were on the order of millimeters for each technique.4 The authors found that dynamic contrast-enhanced CT and MR imaging enabled visualization of vasculature within tumors and surrounding tissues, and examination of vascularization properties including blood flow, blood volume, and permeability.

To guide radiation therapy in real-time, our CT-MRI scanner design consists of a double donut-shaped pair of superconducting electromagnets that forms a regionally uniform 1.5T magnetic field and leaves room for a multi-source interior CT setup and a therapeutic beam generator for radiation therapy (see Figure 1). The CT and therapeutic radiation assembly can be rotated after each exposure. Multimodal images collected during the planning process may be used as prior knowledge. As a result, CT and MRI data required to guide the radiation therapy process can be minimized to a limited number of views through an ROI. These sparse data can be reconstructed in a unified compressive sensing framework. We believe our proposed CT-MRI system could provide more information than MRI alone, and potentially reduce tumor treatment margins and improve survival rates and quality of life.

Figure 1. Rendering of the proposed design for a synchronized computed tomography (CT) and MRI scanner to guide radiation therapy in real-time. Shown are the coil blocks of the superconducting MRI magnet's primary layer (purple) and shielding layer (blue), a cryostat of the superconducting MRI magnet (aqua), longitudinal (green) and transverse (yellow) gradient coils, x-ray sources (red), and x-ray detectors (orange). The coil blocks of the superconducting magnet provide a uniform static magnetic field over a small region of interest, while the gradient coils enable MRI spatial encoding. A multi-source interior tomography scheme is used for CT image reconstruction.

Another application we are investigating for our CT-MRI scanner is the analysis of atherosclerotic plaques. With enhanced features and additional modalities, the system should allow us to investigate many features of vulnerable plaques, including cap thickness, lipid-core size, stenosis, calcification, hemorrhage, elasticity, inflammation, endothelial status, oxidative stress, platelet aggregation, fibrin deposition, enzyme activity, microbial antigens, apoptosis, and angiogenesis. To characterize vulnerable plaques, we will likely need spatial resolution of 50μm and temporal resolution of 20ms. While these targets are not immediately attainable with our current system, they are the goals we are striving to reach as we improve our design.

Additional applications we propose for our system include imaging brain injuries and other brain traumas, and evaluating acute strokes in the emergency room, when time is critical. In 2013, Gonzlez et al. found that the most valuable imaging techniques for selecting patients with severe ischemic strokes caused by anterior circulation occlusions for a specific treatment were non-contrast CT, CT angiography, and diffusion MRI.5 In this situation, quick imaging speed is very important—while image resolution is less so—because the therapeutic window is narrow for excellent prognosis.

In the future, we hope to construct the MRI-CT system we have designed, but this task will not be easy because the scanner is significantly more complicated than multimodal systems currently in laboratories or on the market. We plan to start with proof-of-concept prototypes to show the system's unique capabilities, and then gradually move to product-ready platforms. Our ultimate goal with this work is to make our vision of an 'omni-tomography' imaging system a reality. Omni-tomography involves combining multiple tomographic imaging modalities in one machine and simultaneously capturing images from each modality. The system might include CT, MRI, positron emission tomography, single photon emission computed tomography, and other imaging modalities. We have been performing omni-tomography pilot studies for a few years6 and plan to continue to investigate the system's potential in our future work.

Ge Wang
Department of Biomedical Engineering
Rensselaer Polytechnic Institute (RPI)
Troy, NY

Ge Wang, PhD, is a Clark & Crossan Endowed Chair Professor and the director of the Biomedical Imaging Cluster at RPI.

Feng Liu
School of Information Technology & Electrical Engineering
University of Queensland
Brisbane, Australia
Fenglin Liu
Key Laboratory of Optoelectronic Technology and Systems
Chongqing University
Chongqing, China
Guohua Cao
School of Biomedical Engineering and Sciences
Virginia Tech
Blacksburg, VA
Hao Gao
Department of Mathematics and Computer Science
Emory University
Atlanta, GA
Michael W. Vannier
Department of Radiology
University of Chicago
Chicago, IL

1. G. Wang, F. Liu, F. Liu, G. Cao, H. Gao, M. W. Vannier, Top-level design of the first CT-MRI scanner, 2013. Paper accepted at the Int'l. Meeting on Fully Three-Dimensional Image Reconstruction in Radiology and Nuclear Med., Lake Tahoe, CA, 17 June 2013.
2. G. Wang, H. Yu, Meaning of interior tomography, Phys. Med. Biol., 2013. (In press.)
3. C. R. Rasch, R. J. Steenbakkers, I. Fitton, J. C. Duppen, P. J. Nowak, F. A. Pameijer, A. Eisbruch, J. H. Kaanders, F. Paulsen, M. van Herk, Decreased 3D observer variation with matched CT-MRI, for target delineation in Nasopharynx cancer, Radiation Oncology 5(21), 2010.
4. M. Shukla, A. Kumar, A. Godley, D. Khuntia, Imaging and radiation therapy: current trends and future possibilities, Appl. Radiation Oncology 2(1), p. 6-12, 2013.
5. R. G. Gonzlez, W. A. Copen, P. W. Schaefer, M. H. Lev, S. R. Pomerantz, O. Rapalino, J. W. Chen, The Massachusetts General Hospital acute stroke imaging algorithm: an experience and evidence based approach, J. NeuroInterventional Surgery 5(Suppl 1), p. i7-i12, 2013.
6. G. Wang, J. Zhang, H. Gao, V. Weir, H. Yu, W. Cong, X. Xu, Towards omni-tomography—grand fusion of multiple modalities for simultaneous interior tomography, PLoS ONE 7(6), p. e39700, 2012.