Scott Webb (Host): Innovations in cardiac imaging is our topic today. And my guest, Dr. Himanshu Gupta, is the Director of Cardiac Imaging for the Valley Medical Group, and he's here to explain what cardiac imaging is, the benefits to patients, how AI is being incorporated, and so much more. 

 Welcome to Conversations Like No Other, Heart Care, presented by the Heart and Vascular Institute of Valley Health System in Paramus, New Jersey. Our podcast goes beyond broad everyday cardiac topics to discuss very real and very specific subjects that impact your heart health. We think you'll enjoy our fresh take. I'm Scott Webb. Thanks for listening.

Doctor, it's nice to have you here today. We're going to talk about innovations in cardiac imaging, and I've been excited about this one because I want to hear about the innovations, but I have to be honest though, you know, cardiac imaging sounds pretty high tech. So what does it actually mean for patients and how does it help doctors see and understand the heart better? Tell us how it works.

Himanshu Gupta, MD: That's a very interesting question. Cardiac imaging means looking at the detailed picture of the heart, its structure, its function, as well as understanding the underlying cardiovascular physiology as well as disease state. So what we do is we use different techniques such as echocardiogram, cardiac CT, MRI, or nuclear stress test to comprehensively evaluate structure, function, physiology, as well as the disease state of the heart, so as to make an accurate diagnosis.

Host: Yeah, like I said, it does sound pretty high tech, and I'm sure there's some new uh, let's say non-invasive imaging options. And, you know, I think that when folks hear about heart scans, we're not exactly sure what to expect. So, I'd like to have you talk about some of the newest and innovative things. Exactly what are you doing and how does it work?

Himanshu Gupta, MD: This is a very exciting time in cardiac imaging. There have been tremendous advances both in the hardware, software, and the buzzword of using AI in cardiac imaging. So I would try to dissect out some of these things in a very broad level. And if there are any further questions, I can go in depth.

So from a hardware perspective, obviously with the new GPUs and processing powers, a lot of these scanners that we use are becoming digital and very powerful, meaning that they're able to scan the heart or take the images of the heart in a very high spatial resolution. And that means that we can look at very minute structures, which previously we were not able to visualize as well.

 On the software side, as well as analytical side, we are using artificial intelligence in our processing. And that has also led to combining it with certain physical models. For example, looking at computational fluid dynamics of the arteries of the heart so we can simulate the blood flow within the arteries of the heart, and that can define what is the nature of the blockages in the arteries and whether anything needs to be done. More interestingly, once you combine with other techniques, you can also create digital twin of the heart and what I would say hybrid imaging. So there are a number of advances that are taking place, and obviously it needs to be personalized for an individual.

But in a nutshell, if I was to say both hardware, software, and large analytics are the exciting things that are happening in cardiac imaging.

Host: Yeah, I appreciate that. And as you say, you mentioned there, you know, AI is beginning to play a role and a lot of just great hardware, software options, a lot of tools in the toolbox if you will. Just want to get a sense of it, you said it, when it comes to patients, it's about customizing the treatment plan.

So I'm assuming then that the cardiac imaging helps. So maybe you can take us through that, like, you know, based on the imaging, based on all the tools that you have, how do you develop a individualized plan for the patients?

Himanshu Gupta, MD: Yes. So it starts what I would say a patient journey. So we need to understand why the patient is coming for cardiac imaging, what's the major presentation and what we are trying to evaluate. So based on the indication, we tailor the study for that individual. That would mean evaluating which is the best technique to evaluate the underlying symptoms or condition.

It could be echocardiogram, CT scan, MRI, or stress testing or some other technique. And then, if the patient has been diagnosed with that disease already, then we are trying to understand where the stability of the disease regression or progression. So essentially we are monitoring and then we are quantifying the disease process.

Many of the patients may be in a very early stages of the disease and we therefore need to assess the risk, even in an asymptomatic stage. So literally there are multiple elements, meaning the diagnostic element, monitoring the progress, risk assessment, especially in the early stages, and tailoring the treatment based on what we find, based on our results.

Host: Yeah. Yeah. It certainly sounds like patients are in good hands. And, along those lines, I feel like this is from a lay person, of course, but I feel like these newer imaging technologies are helping you to detect heart problems earlier, maybe even, especially for those who have no real symptoms. Do I have that right? 

Himanshu Gupta, MD: That is absolutely right. So we are really going into what we call precision health. What it means is that we are able to evaluate a person as an individual. A perfect example would be looking at the arteries of the heart. Many a times, patients may not have symptoms of chest pain or even heart attack till later stage of the disease process.

So if we are able to look at the early stages of the disease, we can optimize the treatment. So for some of the techniques that we are using, it would be looking at the calcium deposits, in the arteries. And now more and more we are using what we call quantitative plaque assessment for which we are using very advanced, high resolution CT scan. One of the important things apart from this is that there may be patients who may have disease and have, say, for example, stents or more advanced ages of disease. And there, we are able to use non-invasive techniques, where we are able to look through the stent or other hardwares and we can diagnose a disease, even in the advanced stages with very high accuracy. So the idea being that in the early stage we prevent the disease from progressing and hopefully promote regression of the disease. And in the advanced stage, what we are trying to do is can we diagnose it where patients may need not undergo interventional procedures, or if they're undergoing interventional procedures, the interventionalists know what's the most optimal approach to do it.

Host: Right. Yeah. I think in all areas of medicine, right, it seems to be universal. The common thread is early detection is key, and you give us a sense there of how early detection could, you know, really benefit patients, even providers in a way. And Doctor, I keep thinking about we, we touched on AI earlier and I keep thinking about that.

So I want to come back to that if you don't mind, and just talk about how artificial intelligence is making heart scans more accurate and useful for patients. Maybe you can break it down for us. 

Himanshu Gupta, MD: Definitely. So, we are using artificial intelligence at multiple levels. One is at the level of image acquisition. So the hardware has certain AI algorithms built in that improve the quality of images. As you know, heart is a moving or pulsatile object. So motion is one of the important things we have to account for when we are looking for any object in a static fashion.

So AI algorithms can help in stitching some of these motion artifacts, so this is very important. Secondly, it also helps reduce the radiation exposure. Say for example, in the CT scan or in the MRI, we can abbreviate the study to shorter duration. More importantly, once you have acquired the images which are high quality and de-noise them using certain AI algorithms and got optimal output, then we can use AI algorithms where we can analyze huge data in a rapid fashion. So what it allows is that we are able to get very accurate and consistent and reliable results. Important thing is that we, as physicians are integral element in each and every stage of this process. So we are using AI to complement and in a fashion, in a synergistic fashion so as to provide the best output where you have a human interaction with AI algorithms to provide the best answers for the patient.

Host: Yeah. And I think that's so important to point out, you know, I've done podcasts about non-invasive robotic techniques and doctors always want to point out, surgeons want to point out, you know, that they work with the robot, but the doctors, the surgeons are performing the surgery. And the same thing here you're saying, just clarifying that yes, you're using AI, but it's in conjunction with the expertise, you know, of someone like yourself, a real expert. Right?

Himanshu Gupta, MD: That is absolutely the key because many a times AI is blind in a fashion, meaning that it is, giving a very black and white answer. The nuances of the disease are not as apparent to AI algorithms, and a lot of times it can lead to very wrong directions. So having some human supervision is critical for optimal results and really, I believe for any program to prosper, having a synergistic relationship between AI and human expertise is critical for multiple reasons.

Host: Yeah, that's well said. I love that word synergy and it, it certainly seems appropriate. Just want to give you a chance here at the end, what are you most excited about? I know there's a lot to be excited about today, but in the future of cardiac imaging, what most has you excited?

Himanshu Gupta, MD: There are a number of things that excite me, but what really excites me is the ability to treat or evaluate a person as an individual rather than applying population concepts, meaning we are democratizing the technology where we are diagnosing or evaluating the disease for an individual and we are able to phenotype that person in such a high precision that we can see very early changes that may be happening to prevent the disease from occurring. That obviously means amalgamation of multiple things, but most importantly, amalgamation of AI, huge data and advanced hardware. So I believe the most exciting thing is that at some point, the patients would be owners and able to take reasonably good actions to prevent the disease from happening. So I think the democratization of this, I think is critical for us as a human society, and I am really excited about that.

Host: Yeah, I share your excitement, right. And I'm just a lay person here. I don't get to do all the cool stuff, use all the cool technology, but as you're saying, AI, software, hardware, synergy, just all the buzzwords in medicine, but it really, as I said, feels like patients are in great hands. So thank you so much.

Himanshu Gupta, MD: Thank you.

Host: Thanks for joining us on this episode of Conversations Like No Other, Heart Care. We hope today's discussion on cardiovascular health has been helpful. Remember, your heart is at the center of your overall well-being, and it's never too early to start taking care of it. Don't forget to subscribe and share this episode with loved ones. For more resources on heart health, visit valleyhealth.com/heart.