School of Medicine. I'm here to talk to you today about training pathways in hybrid nuclear CT imaging as part of the ASNIC hybrid imaging workshop. I come to this topic as both the director of a nuclear cardiology lab, as well as a fellowship program director of a cardiology fellowship program. And so these issues of training pathways and competency in nuclear cardiology and specifically hybrid nuclear cardiac CT imaging are very important to me. Here are my disclosures, none of which are relevant to this talk today. So if we look at training guidelines for the incorporation and gain of competency in hybrid nuclear CT imaging, there is no single guideline document that currently addresses this topic in total. What we've done today are taken information from the ACGME program requirements in cardiovascular disease, nuclear medicine, nuclear radiology and diagnostic radiology, incorporated guidelines, including the ASNIC, SNMI, SECT guidelines for SPECT-CT and PET-CT from 2013, the COCATS core training statement, COCATS 4 from ACC. But in addition, a recent document that we were coauthors on from ASNIC and SNMI and other societies about nuclear CT imaging guidelines, which came out in 2022. In addition, we reviewed board guidelines, understanding the competencies that are part of the board exams for American Board of Radiology, the ABIM cardiology exam, ABNM, as well as the certification board in nuclear cardiology and certification board in cardiac community tomography. I mentioned the 2022 hybrid imaging guideline. This was a multi-society effort to try to clarify some of the issues regarding performance, acquisition, interpretation and training in SPECT-CT and PET-CT hybrid nuclear CT imaging being performed in practice today. So taking a step back for when we're talking about cardiac hybrid nuclear CT imaging, there are some guiding principles. The first thing is that CT scans performed as part of these examinations frequently contain clinically relevant cardiac and non-cardiac incidental findings. They have to be reviewed and interpreted. We can't stick our heads in the sand and pretend they weren't acquired or don't exist. However, cardiac hybrid nuclear CT imagers and readers may come from very different training pathways. For example, nuclear medicine, nuclear cardiology, nuclear radiology as part of nuclear radiology fellowship programs and or diagnostic radiology training pathways. And so they may have different levels of expertise and different exposures to hybrid nuclear CT imaging. In addition to variations in training, the equipment that we use in hybrid nuclear CT imaging is highly variable site to site. And they may require unique training and expertise at individual sites. And I'll show some examples from our own institution. So given these variabilities in training, expertise and equipment, that we need to establish a minimal set of standards for training in hybrid nuclear CT cardiac imaging as necessary at both the local institutional and imaging society level. Another piece of background about cardiac hybrid nuclear CT imaging is how common it is. If we think that hybrid imaging or SPECT, SPECT or PET plus CT imaging is our definition, then an unpublished data from a survey that we performed actually two years ago, we found out that over 50% of nuclear cardiology in the labs in the United States have some sort of hybrid equipment. And this is important because hybrid nuclear CT imaging we know increases the accuracy of your imaging. It increases specificity and accuracy through the use of attenuation correction, assessment of coronary artery calcifications, as well as anatomic localization for focal diseases, such as amyloid, sarcoid, et cetera. Some of the data that describes the importance of CT attenuation correction come from examples such as the one I've shown here. Examples such as the one I've shown here from 2016, which shows that CT attenuation correction increases SPECT specificity from 68% to 83% without changing sensitivity. We added to this data in literature in this study, again, going back to 2016, which looked at the added value of the presence or absence of atherosclerotic calcifications on the prediction from SPECT CT imaging on the prediction of obstructive coronary disease in patients with and without perfusion defects. And what we found is that in the absence of atherosclerotic coronary artery calcifications, the positive predictive value of a focal perfusion defect on SPECT CT imaging was only 19%, suggesting that the additional value of SPECT CT lies not only in its ability to improve the nuclear image, but also in our ability to recognize the presence and severity of atherosclerosis in the individual patient. As we talk about the aspects of hybrid nuclear CT imaging, one of the constructs that I like to employ is the concept that there are multiple different aspects to the imaging modality that we need to understand. The first is obviously the nuclear portions of the examination, looking at aspects such as perfusion, left ventricular function, myocardial blood flow, and artifacts. But as nuclear cardiologists, as people who read myocardial perfusion imaging primarily, we also incorporate the aspects of the stress ECG portion in exercise, the duration, symptoms, the presence of ECG ischemia, hemodynamics, heart rate recovery, Duke treadmill scores, et cetera, and in vasodilator stress, similarly, the presence of ischemic ECG changes. But now we're asking to incorporate aspects of the CT, the registration of the images together to ensure that the attenuation correction is performed reliably, the presence or absence of atherosclerotic calcification, also cardiac and thoracic structural abnormalities, such as cardiac structural size, chamber sizes, aortic size, and also presence or absence of pathologic processes in the lungs, mediastinum, upper abdomen, bones, et cetera. All these aspects of the image go into our final interpretation of a hybrid nuclear CT study. Taking that framework further, we understand that different specialties may have different experiences in cardiac hybrid imaging. For example, our nuclear medicine colleagues may have a primary interest in SPECT, for example, with PYP imaging, or with PET in FDG imaging for cancer, infection, and inflammation, and have variable training in CT. Cardiology-trained readers will understand the clinical context about the evaluation of the complementary aspects of stress testing and coronary anatomy, but may need more nuclear cardiology studies and training and may or may not have specific training in cardiac CTA. Radiology colleagues will likely have extensive training in body CT and extracardiac lung pathologies, variable training in cardiac MRI and cardiac CT, and our vascular imaging, and in some cases, exposure to nuclear cardiology and myocardial perfusion imaging, depending on the training program, may be less robust than in other institutions. So in summary, we can incorporate all members from different backgrounds into an overall interpreting team, utilizing their individual experiences, expertise, and training. So in addition to the variability in training experience, let's talk about the variability in hybrid nuclear CT equipment. Let's start with SPECT and understand that there are differences in the detectors, whether it be traditional photomultiplier tubes or newer solid-state detector systems. With PET, similarly, older systems and legacy systems will have photomultiplier tubes, but increasingly, solid-state digital detector PET systems are coming online. These systems can have different fields of view, whether they're wide-field-of-view imaging or cardiac-focused field-of-view imaging, and in particular, the PET whole-body field-of-view imaging with solid-state digital detectors are gonna be fundamentally different in how they image and their accuracy compared to older generation systems using a smaller field of view. Combine that with the variability in our CT systems that are added on to the nuclear cameras. They can have differences in slice thicknesses of the CT from as low as two slices in systems like the Hawkeye Infineon Report or even some newer attenuation correction-only systems up to 160 detectors of advanced PET CT systems. The slice thickness can be anywhere from 0.625 to 10 millimeters, various variances in rotation speed from a quarter of a second up to 29 seconds, providing an average CT over a long period of time, and marked differences in the energy at which the CT scans are obtained. I'll use an example from our laboratory. This is the GE Discovery 570 system using solid-state detector technologies in a 64-slice CT. This is a cardiac-focused CZT system with a high-end CT, sorry, cardiac-focused CZT nuclear system with a high-end CT for anatomic localization as well as attenuation correction. And another example from our laboratory is an older system, an Infineon Hawkeye system, where in addition to having a wide field of view traditional photomultiplier tube detectors, the CT system is a four-slice, slow rotation, fairly high-energy system that does not produce a diagnostic image. So therefore comparing and contrasting those two spec CT systems allows us to illustrate the high variability in systems that are out there and that the individual training requirements may differ from site to site. I talked about the differences in CT acquisition techniques. Here's an example of a CT performed on our 64-slice Alcyon system. And you can see very clearly the diagnostic capability of this system in order to identify both cardiac, mediastinal and extracardiac features in this patient. Let's move to a different system. Another system in our laboratory has an eight-slice CT system that has a governed CT tube current of 30 ma maximum. You can see though, that there is still diagnostic information available on this scan. For example, the degree of atherosclerotic and post-sternotomy changes in this patient, as well as some anatomic features in the thorax and upper abdomen. Let's contrast that with the Hawkeye system I just showed you which is the four-slice, 120 kbp, 23-second rotation. And as you can see here, excuse me, the anatomic information from that CT is not available. And so it doesn't need to be read. So we need to understand at the individual camera level what the capabilities are of our hybrid CT equipment. Within each individual laboratory and within different organizations, there may be different reading models of how studies are interpreted in a hybrid environment depending on the staff, the facility, the capabilities, training, economics, local politics, et cetera. There are several models that we've described. One is where the nuclear cardiologist or nuclear medicine physician, the primary reader, with a radiologist providing backup services for cases only when the nuclear cardiologist, nuclear medicine physician has questions or concern regarding findings on a specific study. The second model is that all CT scans, regardless of pre-overread by a non-CT trained physician, are referred to a radiologist for over-reading. This is obviously the lowest risk model in terms of picking up extra cardiac features and ensuring that all CTs get a dedicated read. A third model might be where a radiologist interacts directly with the nuclear cardiologist, nuclear medicine physician to jointly interpret the nuclear and CT study joint at the time of reading. What model one employs for hybrid nuclear CT imaging depends on multiple factors that I've already described. The last model is that the nuclear cardiologist and nuclear medicine physician who has specific training and background and comfort in interpreting the cardiac and extracardiac findings in the CT interprets both the nuclear and the CT study together. Regardless of the model, collaboration with radiologists or those with specific training in thoracic CT will be necessary for most nuclear cardiology, nuclear medicine providers who do not have prior sufficient diagnostic CT training within the context of a radiology or nuclear medicine residency training program. So let's review what the standards are for training at the current time. If we look at the different pathways that one can train in order to read nuclear cardiology studies, they include nuclear medicine residencies, radiology residencies and nuclear radiology fellowship within the context of a radiology residency, cardiology fellowship and in the general sense for an advanced cardiac imaging fellowship. What I wanna point out to you is that it currently in the nuclear medicine residency, radiology residency and nuclear radiology training, there are no specific guidelines for the duration of nuclear cardiology training. And in many cases, the number of procedures that one must interpret. In so far as the radiology residency, ACGV guidelines covered nuclear cardiology, the single statement in those training guidelines in part that the trainee must have experienced when training in radionuclide imaging of the cardiovascular system. Contrast that to training in cardiac computed tomography. As you can see, there are various aspects of cardiac CT, dedicated cardiac CT that one might encounter during different training pathways. And certainly the non-cardiac CT training by specialty can be highly variable as well. So our general cardiology fellowship trained individuals at this particular time in general, do not have any specific training in non-cardiac CT on certainly no pathway to board certification in that space. We know that advanced cardiac imaging requires an extensive duration of training. If you combine the ECHO nuclear cardiology MRI and CT training for level two training, it's upwards of 17 months, depending on whether these training pathways are served in series or concurrently. And advanced cardiac imaging training, if one wanted to achieve advanced competency in all four modalities could take upwards of 42 months, realizing that most of our trainees focus on one or two level three training methodologies. So as we look to define the standards for practice of nuclear cardiac CT imaging and training, we need to establish a minimal diagnostic skillset for extracardiac findings, a starting point, be able to identify abnormalities in heart other than the myocardium, for example, pericardial effusions, chamber sizes, et cetera, great vessels in the aorta and or anomalies, soft tissues, including the breasts, mediastinal abnormalities, pleural cavity, bony thorax, lung parenchymal abnormalities in the upper abdomen. What we're looking for here is to establish the ability to recognize the difference between normal and abnormal anatomy, defining reasonableness and understanding that different readers are gonna have different experiences and backgrounds, but all should be able to know the difference between what is normal and abnormal, and then refer appropriately for further investigation as appropriate based on an understanding of the basics of normal and abnormal anatomy. So, what are these mechanisms to gain competency? The first are to design training programs that provide cardiac hybrid nuclear CT training. When cardiology, that's gonna include chest CT and attenuation correction CT interpretation. In nuclear medicine, it's gonna require dedicated chest CT exposure, but that's not already integrated into the program as well as attenuation correction CT and cardiac CTA. In radiology, the focus may focus instead on gaining competency in the nuclear aspects of the nuclear cardiology hybrid exam, for spec CT and pet CT. At the local level, each nuclear cardiology lab must have a well-established model to ensure high quality in CT interpretations for extracardiac findings based on the models we've already discussed. And at the societal level, things like this course, the CME devoted towards extracardiac findings, proctored case review, and the further development of guidelines and training statements as the field evolves. So, in summary, cardiac hybrid nuclear CT imaging is becoming the standard of care and that interpretation reporting of extracardiac findings from CT and attenuation correction CT is required. There are many challenges in the creation of standards for these procedures, including variation in equipment, reading models, and reader training and expertise. The models for training in cardiac hybrid nuclear CT imaging will not be uniform and will vary across specialty and site. But minimal standards, competencies, and training pathways are actively in development and courses like this will continue to move the field forward. Thank you for your attention. And I'd like to thank all the members of the Yale nuclear cardiology lab. Thank you.

hybrid nuclear CT imaging

cardiac imaging

training

standards

collaboration