Discussing the value of ultrahigh-resolution CT scanning
September 10, 2019
by Gus Iversen
, Editor in Chief
If you’re a regular reader of HealthCare Business News magazine, the name John Boone might be familiar to you. A couple of years ago we sat down with him to talk about a novel breast scanning CT system he developed alongside colleagues at UC Davis. We recently reached out to him again to discuss another CT innovation — ultrahigh-resolution.
Boone is a researcher, educator, and clinical medical physicist with long-standing interests in CT technology. He is former president of the American Association of Physicists in Medicine (AAPM), and currently chairs the CT subcommittee for this organization. He is also the primary author of the ICRU (International Commission on Radiation Units) Report 87, “Patient dose and image quality assessment in computed tomography”.
HCB News: In April, UC Davis Health announced that it was performing clinical imaging exams using an ultrahigh-res CT scanner. Can you tell us a bit about the technology and how it came to be acquired by the facility?
John Boone: About 2 years ago, I was approached by an executive from (then) Toshiba, about the potential of siting this new high-resolution CT scanner at UC Davis. I was immediately excited about the prospect of evaluating this new technology — and as a leader in the research administration of my academic radiology department, I saw the opportunity for my young academic radiologist colleagues to have an opportunity to evaluate and report on this technology in a series of clinical protocols spanning applications from neuroradiology to musculoskeletal radiology — and that was the primary motivating factor for me in this project.
This CT scanner (which is now the Canon Aquilion Precision) has high-resolution features, which, in my opinion as a medical physicist specialized in CT technology are a game changer for CT imaging — perhaps not for every CT examination, but for a number of imaging applications I believe the high-resolution capabilities will lead to more accurate diagnoses and better patient care. The system has a 40 mm-wide detector (measured at the isocenter, per industry standards), with 160 – 0.25 mm wide detector arrays along the long axis of the scanner (the Z dimension). There are 4 different acquisition modes, which the manufacturer refers to as normal resolution, high resolution, super high-resolution, and ultrahigh-resolution modes. Of course, the size of the detector elements is not the only resolution-limiting factor, and this scanner has 7 different focal spot sizes that can be used, unlike any whole-body clinical CT scanner I have seen in the past. Obviously, the smaller focal spots need to be used with the higher resolution detector modes to actually achieve high-resolution images. The system is capable of reconstructing images with 0.15 mm pixel dimensions, with imaging matrices of 512 x 512, 1024 x 1024, and 2048 x 2048. The MTF, which is the traditional measure of spatial resolution in imaging systems, has a limiting resolution for clinical operation on the order of 3.2 pairs per millimeter, which is consistent with the 150 µm voxel dimensions.
HCB News: What does ultrahigh-res mean in terms of CT scan image quality? Does it open up new capabilities?
JB: We are all aware of the various modalities available in the radiology department for imaging patients and, of course, each of these modalities has its strengths and weaknesses. Traditional resolution CT imaging excels at contrast resolution (the ability to see large lesions with subtle contrast differences) but is not particularly exciting when it comes to spatial resolution (the ability to see small objects) – which for most clinical CT scanners the smallest resolvable element is on the order of 0.50 mm. But with this new CT scanner, we have the ability to resolve 0.15 mm voxels; and the obvious question is, “What human anatomy has spatial features on the scale from 0.15 to 0.5 mm?” The answer to this is the answer to your question about “new capabilities”. We believe that the increase in spatial resolution achievable by this new CT scanner technology will allow meaningful improvement in diagnoses involving lung parenchyma, the anatomy of trabecular bone (in trauma and disease), liver texture, and microvasculature — the latter of which includes diagnoses related to cancer, vascular disease, and a whole host of other pathologies. Thus, the prospective capabilities of this scanner are exciting to me, and through clinical trials we are trying to methodically evaluate which specific areas of CT imaging this new device will advance.
HCB News: Can you provide any examples of how those exceptional images have impacted diagnostics for your patients?
JB: Your readers will recognize that as the voxel dimensions of the CT image get smaller, the spatial resolution increases, but so does the quantum noise. At the same radiation dose levels, fewer X-ray photons traverse a smaller voxel relative to a larger voxel, so consequently, there will be larger quantum noise levels when the smaller voxels are reconstructed. When the Canon scientists presented this technology to me and a group of other medical imaging scientists, of course this was the first question that came to mind. Initially the solution to decrease quantum noise with the high-resolution images was dealt with using iterative reconstruction techniques, which have become the industry standard method for noise suppression in CT over the past decade. I was frankly impressed to find that this manufacturer has also developed CT reconstruction methods designed to reduce noise based upon artificial intelligence techniques — their deep learning reconstruction algorithm. I’m very excited to see how the DLR algorithm reduces noise in the high-resolution images, however this software was only recently FDA approved and is about to be installed on our scanner in the next week or two. For many applications, we were waiting for the DLR algorithm in order to fully assess the system clinically, and we now have our sleeves rolled up to do so. So please ask me this question in a year or so, and I will have a better answer after we have done the research.
HCB News: Operationally, how has training and utilization of the system compared with other conventional CT scanners at UC Davis Health? Is there a learning curve?
JB: This scanner has a large repertoire of options, as you would imagine, and we are still very much on the learning curve in terms of protocol development. We have all become used to the complexity of modern CT scanners, with automatic exposure techniques (tube current modulation), and all the other factors involved in CT protocols. With this scanner, we add to the mix 7 different focal spot sizes, 4 different reconstruction modes, 4 different reconstruction algorithms (each with many flavors), and you can begin to realize how complex this system is. This scanner is a research scanner to some of us, but it is FDA approved and therefore, it is used at our institution for clinical imaging from 7 am to 5 pm. The standard of care imaging comprises the bulk of our clinical investigations comparing high-resolution to low-resolution CT. We really are still at the front end of the learning curve given the complexity of this CT scanner, and I hope over the next few months we will be able to focus on a number of specific clinical protocols and optimize them to achieve the best compromise between spatial resolution and noise. This new high-resolution scanner is certainly a game changer, but realistically it will take a couple of years for my radiologist colleagues to learn how to use this new technology in order to confidently and routinely use this tool to advance patient care.
HCB News: Can you discuss any research you are currently conducting with the use of the system?
JB: As a medical physicist, there are many basic science studies we are planning to do, and these studies are ongoing. We have focused our early clinical investigations on areas of CT imaging where we feel that the spatial resolution capabilities of this new scanner will most impact patient care. Our current priorities are in lung imaging, and I can report that our chest radiologists give 2 thumbs up on high-resolution CT; we are also focusing on musculoskeletal imaging, and have imaged a number of cadaver limbs in various studies which show huge potential; we also feel that neuroradiology applications such as temporal bone imaging is clearly something that this scanner will provide new diagnostic accuracy for. We have a fleet of CT scanners at our medical center, and we are prioritizing the use of this new scanner for a subset of our ambulatory patients to focus on these specific areas.
HCB News: Has anything about the new system been a surprise?
JB: For me personally, there have been huge surprises relating simply to the installation of the scanner. Since this interview is for Healthcare Business News, I feel compelled to mention the following: I wear many hats, but as a clinical medical physicist for 35 years, I have always viewed the delays of equipment installation in the hospital setting as result of bureaucratic inefficiency and indecision. However, as principal investigator of this project, I was intimately involved with most of the steps that were required to install the CT scanner into our clinical CT area. I was so fully impressed at how a thousand people came together in a thousand meetings to expedite the installation of this CT system. I had no idea of the complexity of the process, and am dumbfounded, and grateful to all those who participated in making the siting of this CT scanner at UC Davis possible. From the vice chancellor and CEO of the hospital, to the many administrators, architects, engineers, IT staff, lawyers, X-ray technologists, and so many more that were involved in the necessary discussions and decisions required to install this CT scanner. I have a newfound respect for all those who have a hand in the siting of complex imaging equipment. As a longtime la-di-da full professor in the school of medicine, my advice to similarly gilded colleagues around the country is simple: hug your administrators today!