With the wild growth in Artificial Intelligence, or AI, we wondered how it might be used to improve America’s broken health system. My research led me to Dr. Arman Kilic, a heart surgeon and director of the AI Innovation Center at the Medical University of South Carolina.
The following is a transcript of a report from “Full Measure with Sharyl Attkisson.”
Watch the video by clicking the link at the end of the page.
Sharyl: What was the idea or the concept behind your book?
Dr. Arman Kilic: So the idea was that I think these two fields are starting to merge and intersect, and there’s a real need for each of the respective fields to understand each other. You know, artificial intelligence, the pace at which it’s evolving is astronomical. I would say maybe the most we’ve seen in any kind of prior technology. And as it’s becoming merged more and more into healthcare, I thought there was a need for both groups to understand each other.
Sharyl: I think many Americans would agree, if I say there’s a big sentiment out there, that our healthcare system is just a mess. It’s difficult to get in to see a doctor, it’s difficult to get the services you need, maybe particularly from a specialist. There are long waits, there are shortages of things that we need. What are some of the practical applications of AI that you think could help solve some of these problems?
Kilic: Well, I think there’s a whole variety and a whole spectrum of services that AI can offer. So one is eliminating mundane tasks. So as physicians, you know, we’re spending half of our time documenting notes and doing other things that we don’t necessarily, it’s not a best use of our time. And so AI, right? It’s very good at managing large volumes of texts, speech, audio, video. And then when you talk about shortages, you know, one of the areas probably AI has been most mature right now in medicine is radiology. And if you take a look, you know, if we have a extreme shortage of radiologists and we have to read through a hundred thousand CT scans in a given period of time, and AI can do it in a few minutes, you know, it may help in triaging and then identifying high risk cases or cases where there’s an abnormality, it may also help identify cases where there was a misdiagnosis. So if you have people who are, you know, on hour 15 of their shift and they miss something in a scan, if you have AI running in the background, it may notify them that there was a mistake and allow them to go back and look at that.
Sharyl: Our biggest concern with what AI could do in healthcare would be what— On the negative side?
Kilic: I think the biggest concern on the negative side would be further tarnishing the physician patient relationship. But, I think it all stems down to the core of, you know, how do we preserve the clinician and patient relationship And how do you navigate the use of technology through that relationship?
Sharyl: And then at its best, AI will do what for healthcare, do you think?
Kilic: I think AI can solve some of the problems. I don’t think it’s gonna solve every problem we have in healthcare and we have major problems in healthcare in terms of making the workforce more productive, re-skilling or up-skilling or retraining workers where there may be more needed and where AI can fill some of those gaps. And then again, democratization of expert care. So it shouldn’t matter if you’re in rural America and you know, this state versus an urban area that has a lot of experts co-located in centralized into one area, that we should be able to provide a high level of care no matter where you are in the country.
Sharyl: You’re optimistic?
Kilic: Yeah. Yeah.
Sharyl (on-camera): Google recently announced it’s training its AI model with 300 million audio samples, including coughs and sniffles to train it to diagnose illnesses.
Watch video here.
NOW AVAILABLE: FOLLOW THE $CIENCE
Order Sharyl Attkisson’s New Bestseller!
