Omni-Tomography: An All-in-One Modality?

What if you could combine your CT and MRI machines, obtaining simultaneous and in vivo images? You’re one step closer, thanks to researchers like Ge Wang, PhD.

Wang, director of the Biomedical Imaging Cluster at Rensselaer Polytechnic Institute, and a group of international collaborators developed a high level design for a combined scanner that does just that. They will present their results at the Fully Three Dimensional Image Reconstruction in Radiology and Nuclear Medicine conference this June in California.

Omni-tomography is Wang’s terminology for not only combining imaging modalities in one machine, but also simultaneously providing images from each.

“That is what I call all-in-one and all-at-once, performing scans instantaneously within one gantry. That’s my dream,” Wang said. Putting CT and MRI capabilities into one machine is one type of omni-tomography, and will provide major diagnostic and interventional benefits for patients, he said, like cardiac imaging and therapeutic guidance, guiding procedures like heart valve replacement, and documenting the delivery of drugs or stem cells.

How is it possible? Interior tomography.

“Interior tomography allows exact reconstruction from less data which are highly truncated,” Wang said. The process uses a narrower beam on the internal region of interest (ROI), with the added benefit of decreased radiation to the patient. It also gives improved contrast resolution due to fewer scattered photons.

“Previously computed tomography targeted either global reconstruction - the whole body cross-section - or approximate local reconstruction which can be significantly distorted,” he said. “The change is interior tomography. Now we target a smaller region and yet are able to produce accurate and stable local reconstruction.” Wang and his collaborators have been working on interior tomography for five years, showing that local reconstruction can be theoretically exact. “That wasn’t known before. This is what we contributed to the imaging field. That’s important because only when you have theoretically exact reconstruction can you perform quantitative analysis for diagnostic information.”

Wang noted that interior tomography is not limited to one modality, he said. “Most imaging modalities can be modified to target an ROI for accurate reconstruction.” A difference is in how it measures. “In interior tomography, you don’t need a direct measurement of an ROI, only indirect measurement,” he said.

MRI normally uses a universally large, homogenous magnetic field. “We propose using a small homogenous magnetic field,” Wang said. “We proved that it works in a PLoS One paper published last year.”

Building a new machine

[[{"type":"media","view_mode":"media_crop","fid":"12287","attributes":{"alt":"","class":"media-image media-image-left","id":"media_crop_3218353789248","media_crop_h":"0","media_crop_image_style":"-1","media_crop_instance":"430","media_crop_rotate":"0","media_crop_scale_h":"0","media_crop_scale_w":"0","media_crop_w":"0","media_crop_x":"0","media_crop_y":"0","style":"margin: 5px; float: left;","title":"Rendering of the proposed CT-MRI scanner","typeof":"foaf:Image"}}]]While current MRI and CT scanners are large, Wang said that by using interior tomography, a combined machine is feasible when each of the two modalities is made slimmer and more compact, because the detectors are smaller, and magnets and coils are more focused. By using a narrow beam for a small detector piece instead of a wide beam for a large detector panel, the detectors cost less and take up less space. Not only would the machines ultimately be less expensive to produce, but multiple detectors fit in the machine for parallel data acquisition.

“Truly hybrid imaging is possible. You now fit different modalities together,” he said. As for combining CT and MRI and having them work simultaneously, “the feasibility is recognized,” he said. By using interior tomography, MRI can use a weaker magnetic field, allowing for compatibility with CT scanners, and the machines can be physically fused without interference.

Wang is looking for funding to build a prototype. “If we have sufficient funding, we can produce a good prototype in five years.”

While there have been some concerns about the economic feasibility of combining the machines, Wang dismissed them. “When computers were first invented, some said they were not very useable and only needed a small number,” he said. “Omni-tomography is a new type of imaging. The machine can address some unique issues and the cost will come down. Only in omni-tomography you see in vivo happening comprehensively and simultaneously. It’s a dream machine.”

[[{"type":"media","view_mode":"media_crop","fid":"12285","attributes":{"alt":"","class":"media-image","id":"media_crop_6073184195015","media_crop_h":"0","media_crop_image_style":"-1","media_crop_instance":"429","media_crop_rotate":"0","media_crop_scale_h":"0","media_crop_scale_w":"0","media_crop_w":"0","media_crop_x":"0","media_crop_y":"0","title":" ","typeof":"foaf:Image"}}]]

Interior reconstruction with the tensor framelet (TF) from 27 projection views. (top left) The image phantom with a central 200×200 ROI and a coarse-resolution version of the image outside of the ROI, (top middle) and (bottom middle) the iimage reconstructed using FBP and TF respectively, (bottom left) the true ROI image, (top right) and (bottom right) the error images associated with (top middle)  and (bottom middle) respectively. Images courtesy Hao Gao, PhD, and Ge Wang, PhD.