Michelle Moyal, DVM (Host): Welcome to the Cornell Veterinary Podcast, where we deep dive into the discovery, care, learning that happens at Cornell University's College of Veterinary Medicine. I am Dr. Michelle Moyal, the Chief of the Primary Care Surgery Service and Knowledge Seeker, not to be confused with Quidditch Seeker, if you know, you know, and general extrovert, and I am very excited for today's guest.

Today's guest is Dr. Parminder Basran. He is a Radiation Oncology Physicist. Three amazing words put together, by the way. And Associate Research Professor in the Department of Clinical Sciences. Prior to coming to the college, he was a Radiation Physicist working in human medicine using artificial intelligence to analyze radiology images, a field known as Radiomics. He has started the quote, AI in veterinary medicine group at Cornell; amazing, gathering faculty members who apply these methods in their veterinary research. He is also launching the symposium on artificial intelligence and veterinary medicine, um, AKA Saivm in April of 2024.

And listeners, I'm hoping we will see you all there. Welcome to the podcast. Thank you for joining us.

Parminder Basran, PhD: Thank you for having me. This is exciting.

Host: Oh, I'm very excited. Generally, amazing researchers, and scientists like you allow me to bombard you with questions that may make sense to other researchers and, not make any sense to some of our students and future researchers and scientists listening. And so the goal is that everyone, gets to listen and to be as excited about physics and AI as you are, and I will be. So, would you mind, telling us a bit about your professional and personal journey? Like, how did you get to where you are right now?

Parminder Basran, PhD: I kind of like to think of my journey as being sort of a series of unfortunate misdirections. I'm a second generation immigrant and I grew up in small and remote places in Canada. And I kind of stumbled and wandered into physics and math largely because I was pretty much horrible at everything else. And then through a variety of misdirections, I found myself landing into this field called medical physics. Medical physics is this field where we use the principles of physics to help people and animals, diagnose their diseases, treat their diseases, primarily focused on things like medical imaging and radiation therapy.

So after I, did my undergraduate and graduate work, did a bit of training. I worked in a variety of centers in Canada. I worked there for about eight years in Toronto, Calgary, and then I moved to Victoria, British Columbia as a Senior Medical Physicist. And one of my jobs there was actually kind of drifting away from daily patient care and more into management of radiation oncology systems.

We manage the world's largest database of radiation oncology, servicing a population of about 6 million people. And I really loved having that work, then about four and a half years later, I had this amazing opportunity to come to Cornell University and put some of that knowledge that I gained in clinical practice to practice here in the Vet school .

I guess one of the things I could mention is that, you know I really loved my job when I was Canada doing medical physics, clinical medical physics and helping people receive cancer care. But one of the things that I wasn't able to do was, was a lot of research.

And so, coming to Cornell really afforded me an opportunity to sort of sink my teeth into research and, find ways to leverage all the knowledge I gained in, in clinical practice here in the vet school.

Host: Oh, I love that. I love that. I love that you call it a series of misdirections because I really think a lot of us who find ourselves in the vet school or in veterinary medicine, have the same experience and you've actually touched upon a few things that I would love to get to. So one, to our neighbors in the north. One of your own is here. Here we are, we're chatting, the other thing I wanted to touch upon, which actually really resonates with me. So I am a first generation American and, I very much appreciated that you shared that with us and our audience. And, I think there's a lot of talk about diversity and diversity in STEM and, all of these things. Why do you think it's important for people from different backgrounds and places and lifestyles to be in science, to be in a position like yours?

Parminder Basran, PhD: Yeah. I love these kinds of questions. I love this kind of space because one of my real sort of ethos is, is improving access to care, globally and whether that means you're providing care to a human or to an animal; it's a really important part of my personal sort of ethos and the things that I like to work on. And I strongly believe that, when it comes to empowering, think a lot of a way we can improve the quality of life for humans and animals is by empowering people from different cultures and different types of countries and, different backgrounds, to be involved in things that they never really would have thought of being involved in. When you think of like certain professions and certain fields, you may have certain stereotypes of what that kind of person or what kinds of identities, those people might have. But I, firmly believe that the more weirder groups of people you can get and more people from different backgrounds you can get into these spaces; I think, you really enrich the environment, where people can work. And there's a couple of important considerations there. You know, the first thing, you is just the fact that, you seeing people of different colour or a different accent or a different space being in those places where you wouldn't traditionally see a lot of people of color or whatever, really helps sort of break down barriers for younger people. We see this a lot where we are in different parts in society, in STEM, and other fields. So I think, this is kind of a big deal for me. I love it when people kind of approach and ask me what is it like to be a medical physicist? people from different countries and different cultures and show them there's a path for anyone be successful in STEM.

Host: Yes! Oh, I love that answer. And representation is a big part of the joy, I think, for me being in an academic institution, and I feel like that kind of joy just radiates off of you, and so I very much appreciate that. Amazing, and you heard it here, folks. Cornell is keeping it weird, and we are happy about it.

So I love that. Because I think different perspectives and different thought processes, whether or not it's in the research realm. So when I took physics back in the day and I struggled to figure out how long it would take for a football to land when I threw it at a certain angle; you have gone into this amazing direction, helping people, helping animals, it's just incredible what you're doing.

I was just curious because you kind of told us that you came to Cornell and you get to do all this amazing research. What is it like to go from human medicine to veterinary medicine, like, is it really similar? Is it very different?

Parminder Basran, PhD: I had very little knowledge about what veterinary medicine was all about before I came and visited Cornell. I mean, you kind of read textbooks and you know people that have animals and you have experiences working with vets in sort of communities, if you have animals or pets; but to really know what it's like to work at a veterinary center, something like Cornell, is something I had no idea of.

When we were deciding to come out here, I had this sort of wonderful opportunity to meet some sort of seasoned faculty here at Cornell, people like Dr. Margie McEntee and Dr. Thompson, and I also had a chance to meet a bunch of other folks just here on the Cornell campus, and I think the thing that really struck me was there was a couple of really interesting things. The first thing was just that, I kind of saw the kinds of challenges and problems that folks had here in veterinary centers and how similar and different they are to the kinds of challenges we have in human medicine. So, for example, if you take a look at the health information system, like you have a patient chart, right? And, I know what that looks like in human medicine. I've, dug my teeth into solving problems of workflow efficiencies and, and how to manage these systems and how to best use them.

What are the best practices and what are sort of ways you can do quality control and quality assurance on these very complicated systems. And to think that these similar systems exist in veterinary medicine, but there's so many really interesting questions in relation to all those things that I just mentioned, like, how do we ensure that these health information systems have the adequate quality assurance and quality control measures, or how do we best leverage these technologies, for the kinds of practices that people have?

There's a lot of nuances and differences, between the way humans practice medicine, and the way veterinarians do. So I, found that space, really exciting and interesting. And I guess it kind of fit in well with this, I didn't know it at the time, but this phrase One Health was something that I always resonated with. The idea of helping humans, animals, and, keeping the earth sustainable. These are all sort of big picture things that we want to try to bridge together. And I kind of really saw that here in Cornell, in the vet school. It's a little more challenging to see that kind of headspace in human hospitals, because we're always really stressed on trying to ensure that patients get treated very efficiently and safely.

So that was one of the things that I saw, like there's a lot of similarities, a lot of differences, and there's a lot of really great questions that are easily transferable from one space to another.

Host: I love that. And there are even differences between how I might utilize a record or write a record as a generalist than maybe like a specialist. So I love that that's a focus. And I think a lot of our audience members, whether or not they're in the veterinary community or not, sometimes don't even think, they just think that a veterinary institution is just about animals, right?

But the truth of the matter is, that a lot of this translates to the human side and a lot of us do this research, one, to enhance the relationship that humans and animals have, but two, that it can really advance human medicine, human health, and I just love that.

Parminder Basran, PhD: Yeah, the whole space of like epidemiology and the connection between like outbreaks of diseases and things like that, you know, there's just so much interconnectedness between the health of animals and humans. That we have so much more to learn by learning with each other, as opposed to, just ourselves in our own little silos.

Host: Absolutely. And, we might've gone through some of that during this previous pandemic and that whole thing that happened. And, and now we're seeing kind of this canine respiratory illness and stuff, but I don't want to get too deep into that because there are so many questions I have for you.

Okay. So artificial intelligence, AI. We hear it thrown around all the time. I'll admit, I heard AI and I thought about the movie Terminator. Maybe that's not where most people should go. And so when you talk about it, what are you referring to? Like, how is AI used in medicine?

Parminder Basran, PhD: Today, I think when people are thinking of AI and how it might be used in medicine, I think it's fair to say that it's mostly in this, what we call narrow or weak IA, as opposed to sort of this general IA, where you expect an AI system to come up with a diagnostic plan and a workup and everything else. Like most of the algorithms that we see being deployed in human medicine and some and even veterinary medicine are, as I mentioned, something we call narrow AI where you kind of have a very specific set of tasks or problems that you want to try to solve, and you need to find a way to automate or to make that decision making process a lot more efficient.

So, you take, for example, the idea of a radiologist reviewing an x ray image. A radiologist can spend, you know, a certain amount of time to analyze that data, but you can also do it in a fraction of a second with a deep learning type of AI algorithm.

And that information then goes to the radiologist and helps them with a decision support. So I think those are some examples of how we see AI. They're not sort of like this broad, Terminator kind of thing. They're very specific problems, that people need to solve that help them make the next step in their work.

And nowadays there are literally thousands of these algorithms that are being operationalized in healthcare, in large centers. So, we're seeing like these sort of small bits of AI contributing to different parts of people's work, to make them more efficient in delivering healthcare.

Host: I did some continuing education in this not too long ago and they were talking about an AI program for radiology or like reading x-rays for those of us who are used to that or other images, right? And so, I want to make sure we understand, what they had spoken about at the time was, you know, if you get ten radiologists to read some x-rays, you get ten different answers, and so, AI may be a way to streamline this by using an algorithm to help kind of eliminate some possibilities. So what I'm hearing here, which I think is really important, AI is not being used to replace people.

It is being used to enhance what doctors do. They're like, listen, it takes a few seconds. Here's what this algorithm is showing. Now you as an experienced clinician who has studied radiology uses that info to make your diagnosis or your report, things like that.

Parminder Basran, PhD: Yeah, that's a hundred percent right. A lot of people think that AI is going to replace humans. And my take on that right now is just that I don't see that happening at all. In fact, I still see there being a really important need to have warm bodies do some of this important work.

AI can help you work more efficiently and work smarter. But it's not going to replace the kind of comprehensive skills that a veterinarian has. I think of the fact that, we have these projects where veterinarians would go out to these large dairy farms and do like, assessments on these dairy cows. Well, when they go to these farms, even though they might be doing a very specific thing, like doing some kind of udder checks or teat health score analysis or something along those lines, their brains are collecting a lot of different types of information that could have different outcomes.

So if the algorithm is only doing a detection of teat shape, well, that's all the information you're going to get. But if you have the veterinarian out there, the expert out there, they're able to provide context to a lot of the AI that might be in use. So, there's still a lot of work that needs to get done to make AI more into this sort of general AI that we kind of all have in our minds as being this sort of romantic notion of what artificial intelligence is.

So. Yeah, it's not a panacea, but it can be a decision support tool.

Host: Gotcha. So, okay, I'll save that for a movie, thinking about that for movies, but just great support for us here making human medicine and veterinary medicine even better. And that's what we want to advance health. So we talked about kind of the radiology side, what kind of applications are you thinking about for the future? Like, what do you think the potential for AI could be in the future?

Parminder Basran, PhD: There's just so many and I can list off a couple. I think, the image analysis thing is a really big area of potential exploration. Any kind of imaging, whether that's an X-ray or a CT or, infrared imaging, that kind of analysis, classification of images and so forth.

There's the whole scope of diagnostics and decision support that I mentioned, like how do you use this information to help support what the next step might be. Another really important field of AI is in the background of drug discovery. So, if you take, for example, doing omics profiles or proteomics or, you know, DNA analysis of large data sets, you need really powerful algorithms to find connections on how you might be able to use some of that information to improve the development of drugs or other kinds of treatments.

And that kind of happens more sort of on the commercial side. And then the other really big one is, is just the whole idea of predictive modeling, can you collect a variety of features from an animal at time point A, and then use that to predict what the risks of certain diseases or certain conditions might be in the future. I could go on and on about what the potential applications might be. Those are just some of them.

Host: Neat. So even for someone, you know, I always try to think about how I feel in my position as a primary care doctor, right? So for all our new doctors that are starting out and veterinarians are in need right now, maybe one day as a new doctor, they could use AI to their advantage in just thinking about certain disease processes that could be happening in their patients and again, that's smaller scale and on a large scale, you have these companies that could say, let me look at the trends when it comes to this treatment or this drug and make these connections that could suddenly lead to new treatments, new uses for drugs, am I catching what you're throwing out?

Parminder Basran, PhD: Yeah, exactly. And, veterinarians can use AI right now. Some of the things we don't really think of are sort of the mundane tasks that we have to do on a day to day basis. One may not call it mundane, but often, people have to do a lot of dictation and generate reports or findings or something along those lines.

The most common way to capture people's information is to use sort of free text and sort of blurb things out and try to collect that onto a piece of paper. And then that paper gets shipped out, for storage or use and elsewhere. But today, we can leverage AI to sort of standardize reporting, for example.

So you could take free texts, just dictate that into something, copy and paste it into chat GPT, which you might have trained to say, make sure you generate a standardized report that looks like X, Y, and Z and then spit out the report, based on all of that work. So instead of having to spend 20 minutes of your life pointing and clicking and trying to get your documents to look the way they want to look; you could just leverage ChatGPT, or bert, or some other AI tool that exists right now to help with all of that. So, I think there's so many ways we could actually leverage that even just today that we don't really think about. I often think that we should have maybe like a crash course on just what are some like softball things that we can use for AI to help our lives in our daily work.

Host: Yeah, after many years in emergency medicine and many, many years in general practice, let me tell you, I love some mundane. I will take some routine days and if anybody has seen me typing, if you have, don't comment on the podcast, stars or, or anything like that, because I can't really type great. Something like that would definitely help someone like me. So, that's amazing. But for the students, please write everything out for right now. ChatGBT can wait. That's so cool. Okay, that's like just really neat and it gives me a better understanding and I'm hoping some of our listeners a better understanding of what it is because when I hear AI, I think it's like, far off again, like robots doing work. It's such an odd image. And, and again, that shows that I grew up in the eighties watching some very interesting sci fi movies. You are going to be hosting a brand new symposium. And I mentioned this earlier, the Symposium on Artificial Intelligence and Veterinary Medicine, amazing, um, in April. So I guess my question is why? Like what, made you decide do this?

Parminder Basran, PhD: There's just so many important questions that need to be answered or discussed, when it comes to artificial intelligence and veterinary medicine. I think there are a lot of great conversations happening in human health care. We're starting to see a lot of guidance documents issued through different governmental bodies, whether it's the EU or here in the U.S. But there's not a lot of discussion about all of these things in veterinary AI. Things like, who owns the data? Who gets to decide what data is public or private? What does ethical AI look like in veterinary medicine? That's going to be a tremendous issue for us, especially when it comes to considerations where the AI technology could be used to make really critical decisions about for example, euthanasia or not. So, who's responsible for that kind of decision making process. And so there hasn't been a lot of structured conversations about these topics.

And so one of our goals is to have a forum where we have an opportunity to have very structured conversations about these kinds of tough questions. But also to explore the different ways that we could use AI in veterinary medicine. And so the four pillars of our symposium are to explore AI in the context of One Health.

So it's sort of comparative medicine, you know, humans and animals, companion animals, specifically, looking at livestock health. And then the fourth pillar is, population medicine. And so we're going to have four different branches of talks that will focus on those different elements.

So again, connecting them all together this sort of this One Health paradigm. So this we feel is going to be an exciting, groundbreaking event where it's the first time ever where we'll have experts from all over the world, sharing their thoughts on what AI should look like in veterinary medicine.

Host: Wow. And what do you hope that those attendees get out of this? What do you hope they leave with? Is there a message? Is there just some thoughts?

Parminder Basran, PhD: A couple of different things. I think one of the things we want is to really start having people think about tangible projects, research ideas that they can take away from this symposium. What are the kinds of things that we really should be looking at? Or what are the big holes in in AI, in veterinary medicine. Should we be focusing on data collection principles, or should we be focusing on ethics?

I want to help identify what we feel are sort of high priority things that this community should be looking at, and maybe direct our energy towards those things. And then of course there's a whole sort of scientific part of any symposium where people get to connect and share ideas and potential collaborate. And then there's the fun part, where you can just get people together and have a nice meal and talk shop in a warm, safe, comfortable, environment.

Host: Yes, everything good happens over snacks. That's my motto. I love that. And you actually mentioned something that I was just going to ask you about the ethics portion. So as you talk about the implications or kind of broad applications of AI, what the ethical obligations we might have on this side of the world. If any, like, maybe we haven't gotten to a point where we have to consider that.

Parminder Basran, PhD: We really ought to start thinking about these things as a community. There was a really great paper that was published a couple of months ago, that was in the American Journal of Veterinary Radiology and Ultrasound, the VRU Journal. There was a supplement, an AI supplement. In that supplement, there was an excellent article on ethics of AI. So I would encourage anyone that's really interested in it to dig up a copy and go through that. And the authors went through and described in a lot of detail, what are the important things to be thinking about? And I mentioned some of those earlier, things like, what impact does this have on critical decisions like euthanasia? Who's responsible for a decision that's made when people use AI? All of these things we really do need to start thinking about because it's going to come very fast.

That's the thing I'm worried about is that, certainly a lot of commercial parties are going to really try to push these technologies to veterinarians who we know are really very busy and sometimes overworked and stressed. And so AI can be a solution for managing a lot of those workplace related challenges.

And if that does happen, if people start incorporating more AI into their clinical practice, we have to be really mindful of what the consequences are. We have to be very mindful of the data that's going into these AI algorithms and whether there's systemic biases in them. We've seen this in human medicine, ad nauseum, so many instances of how really bad data can do all sorts of horrible things like discriminate against certain populations or just not work very well or are not generalizable enough to work in different types of communities and practices. Those are the kinds of questions we need to tackle in veterinary medicine as well.

Host: Yeah. Amazing. And it's, again, it sounds like the speed at which some of these programs come out can be so quick, we don't have the time to digest and take a second. And as a practitioner, I agree. I might say, Ooh, that makes my work so much easier. Let's just get that, without thinking through some of the implications, and we might not know. We might not even know what to think about. So I very much appreciate that comment.

So I'm just curious, and I've asked this of some of our guests before, when you're in your lab and when you're doing work, I need to know, are you listening to music? If so, what is it?

Parminder Basran, PhD: Absolutely. I have a very broad range of musical interests.

Host: He's not committing, everybody. He's not committing. I'm going to need a group.

Parminder Basran, PhD: I listen to a lot of stuff. I was just listening to some old school Busta Rhymes, just a few minutes ago. What's on my playlist right now? It's all over the place. I listen to a lot of genres. The only one that I don't really have a lot of affinity for is the twangy country music stuff. I'm just not really into that stuff. But everything else I'm absolutely into.

Host: Feel free to make your comments about you may like all music, but everybody knows that they have a soundtrack or something, soundtrack that they exercise to, or they do their work to, they sleep to, or they have in the background when they're doing stuff. We all have them. We all have our preferences. And our students have introduced me to a lot of different kinds of music, that are really interesting. They make me feel really, really current. Makes me sound not hip. And so I guess, what is on the horizon for you that's really exciting to you, I know you're doing all this amazing research, but what's one thing where you're like, yes, I get to do this. Or I'm trying to do this.

Parminder Basran, PhD: Not to toot the Cornell horn too much, but I think we have a really unique opportunity here at this university. We have some amazing computing scientists just down the road, that are just doing some amazing, things. and then we have, Cornell Tech, this sort of bright star in Manhattan, that's doing all sorts of exciting work in AI and and then we have Weill Cornell Medical Center. We have the connection there. And then when you couple that with the fact that our Cornell University Hospital has some incredible resources for data, it really makes me super excited about the opportunities that we have here and the kinds of things we could do in artificial intelligence, the kinds of models that can be developed and the kinds of training that we can offer to students, vet students, or students from different parts of the campus. I think there's just so much opportunity here at Cornell that gets me really excited and I think there's a lot of things we can do to help push the envelope a little further along and develop Cornell into a vet AI powerhouse.

I think, that's a totally doable kind of thing. So that's one thing that I would be really excited in exploring. Another thing is, is that I'm a Radiation Oncology Physicist by trade and we are getting a new linear accelerator. So I'm super pumped about exploring that technology and how we might be able to find clever ways of treating different kinds of diseases in animals and then maybe leveraging that for use in human medicine as well. So, there's a lot of, a lot of exciting things going on right now.

Host: Limitless possibilities. So for some, that brings excitement. For me, just a little bit of anxiety. That's a lot of choices. I don't know if I could do it, but that's amazing. And, what? Cornell has a medical school? Who knew? Just kidding, everyone. But just quickly for those of us because I want people who have never heard of a linear accelerator to know what it is. In the plainest of language, what is it?

Parminder Basran, PhD: It's a beam that shoots photons at cancer.

Host: In the attempt to get rid of it, remove it.

Parminder Basran, PhD: Get rid of it, to cure it, to treat it, to help patients.

Host: Amazing. And we're getting a new one.

Parminder Basran, PhD: We're getting a brand new one, it's going to be the coolest linear accelerator in all of our county.

Host: You hear that Canada. Yeah, we're talking to you. So, okay, a couple of quick lightning round questions before I ask you, one final question because I don't want to keep you from all this important research you are doing. Favorite movie?

Parminder Basran, PhD: Oh, City of the Lost Children. It's a French movie. It's brilliant. Yeah, it's my favorite movie.

Host: Okay, everyone, you heard it here. Sugar or salt? An acceptable answer would be you can dip your french fries into ice cream if you want.

Parminder Basran, PhD: I'll have to say salt.

Host: I'm okay with that. I support it. But really, I'm both. And your favorite, so we have a lot of listeners who are not from Ithaca or not familiar with Ithaca. Obviously, there's a lot of beautiful things here. Favorite thing about Ithaca.

Parminder Basran, PhD: Favorite thing about Ithaca. I love the four seasons. Autumn is just marvelous here.

Host: it really is breathtaking here. I'm not going to lie, I'm not just trying to sell it between our waterfalls, but like leaves changing out here is like ridiculous. Plus the Apple Festival happens during that time. So there's that. So just a, final, question for you. Is there something that I maybe should have asked but didn't?

What do you want our listeners to know? What do you want them, if there's one thing you want, it doesn't matter what aspect of veterinary medicine, science, what aspect of training they're in, what would you like our listeners to know?

Parminder Basran, PhD: I always like to tell people don't be afraid to step outside your comfort zone. And you'll be surprised how welcomed it might be for you to be in some other gray zone. Like, I think the coolest things that happen in people's lives are when they take a few risks and are not sure what might happen to them.

And, I kind of look at my own trajectory as being a perfect example of that; where I've stepped way outside my comfort zone, but I'm having so much fun and really enjoying it. So that's my little word of wisdom I will leave with you.

Host: I love that. The misdirections. They are opportunities, even if they don't feel like opportunities at the time. But it's absolutely true. And Veterinary Medicine and Cornell welcomes all of the people that are experiencing those misdirections because you could do something profound somewhere else. And I love that. Dr. Basran, you have been an incredible guest. Thank you so much for joining me today on the Cornell Veterinary Podcast and I hope everybody stays tuned, for future episodes and we will talk to you soon.