'THE ULTIMATE PHYSICIAN EXTENDER': AI expands doctors' capabilities in patient care

[00:00:04]Historically, healthcare has been somewhat slow to adopt new technologies, and we are seeing the exact opposite with AI, especially generative AI and the latest technologies that you're seeing in these large language models.

[00:00:18]We're adopting this technology about two times as fast as any other industry in the American economy. So, we are seeing the impact across the full spectrum of both discovery, the research, bringing new medicines and treatments out of the laboratory and in delivery, making sure that every American has affordable, accessible, safe, high-quality care with the physician and provider workforce that we've got.

[00:00:47]>> That's Dr. Peter Najar of Johns Hopkins with me earlier this week as we take a deep dive into AI and its acceleration across healthcare. All week long, the growing technology also assisting sports medicine doctors, helping to diagnose injuries faster and giving athletes and trainers personalized rehabilitation and training regimens. Joining me now is Sports Medicine Institute at the Hospital for Special Surgery orthopedic surgeon and chief emeritus, Dr. Andrew D. Pearl. Dr. Pearl, great to have you.

[00:01:14]>> Thank you, Maria. It's great to be here.

[00:01:15]>> Thank you so much for being here. So, researchers are talking about creating digital twins of joints using AI. Tell us how AI has changed sports medicine and about those digital twins.

[00:01:27]>> Well, I think AI already has been introduced into medicine and we're all using it. As Dr. Najar said, it's making us more efficient practitioners. It's the ultimate physician extender.

[00:01:39]And so, it's allowing us to deliver care in a much more efficient way in terms of our documentation, in terms of our literature reviews, and then in terms of tracking patients.

[00:01:49]>> I understand documentation, for sure.

[00:01:51]And I also understand how AI can read perhaps a scan better than or or adding to the help for a doctor. But, what about this digital twins where you're actually creating a digital twin of a person's joint? How does that work?

[00:02:06]>> Well, I think that's kind of the next iteration of taking these technologies and bringing them to them together.

[00:02:11]Digital twins are virtual replicas of a physical asset used to run simulations, and it's been used in manufacturing for years, but it's at its nascency in medicine. It's been used in cardiology.

[00:02:23]There was a seminal paper that was just released this spring talking about the use of digital twins to guide ablation treatments for arrhythmias, and we're starting to use it in orthopedics where we can create digital twins of somebody's joint and run simulations to understand how that joint might behave under different loads or with different sporting activities.

[00:02:43]>> That's incredible that to have the digital twin to actually look at that to see what's affecting your knee, your your shoulder, whatever. What are you seeing in terms of how that's being used in real practice?

[00:02:54]>> Yeah, the case studies for it or the case use cases for it are for surgical planning.

[00:03:01]I'm a knee surgeon, and so we always have a number of different surgical options for a patient. Now, we can run simulations and identify the best surgical option and the best surgical technique for that option.

[00:03:14]But, we also see it as a way to predict injury going forward. For example, um lots of people tear their ACL. Um you have people who play the same sport, same exposure. Some people will tear their ACL, some people won't. It's always been a mystery as to why certain people were more susceptible to ACL tears. Now, with digital twins, we can simulate those sporting activities and their impact on a knee and maybe predict who might be at risk for an ACL tear, and we're starting to do that.

[00:03:44]>> that. And what do you do in terms of rehab once you have that tear?

[00:03:48]>> Well, um usually ACL tears, if you do have an ACL tear, unfortunately results in a surgical intervention, at least for young active people, and then there's a prolonged rehab.

[00:03:59]But, we're starting to use AI to understand what risk factors make you susceptible to further injury in that knee, or in the other knee, which is very common.

[00:04:09]>> Really fascinating. We're also seeing AI used by trainers to zone in on specific movements and workouts to enhance performance, right? Tell us about that.

[00:04:17]Is that the direction that you see AI taking?

[00:04:20]>> Absolutely. So, with AI, we can use what's called markless motion traction motion trackers, and we can understand how somebody moves and really quantify how somebody moves. This really helps trainers and doctors understand movement patterns and helps them optimize movement patterns. And even patients can interact with these systems and see how they're moving and do corrective activities to improve their movement patterns.

[00:04:47]>> Yeah. Meanwhile, earlier this year, HSS implemented its own agentic AI platform to transform patient care and person wait times, accelerate a path to treatment. Tell us about that. What are you doing specifically at HSS beyond this?

[00:05:00]>> At HSS, we've got great system engineers that have pulled our data.

[00:05:06]You know, AI is great for synthesizing and analyzing large data systems.

[00:05:13]Hospitals traditionally had paper charts and things like that. Now, we have electronic medical records, we have MRI, we have all this digital information.

[00:05:21]But, you have to pull it, you have to get it in a central repository, you have to have it in a data lake, and you have to make it accessible. And so, we've had terrific system engineers who have put that together for us. We're relatively small hospital focused on musculoskeletal care. So, we can really access that data in very unique ways now.

[00:05:38]>> Absolutely. And the Hospital for Special Surgery has been the leader in its space for so long. To add AI on top of that and this new implementation and what you're talking about digital twins must be at night and day in terms of like 10 years ago. Is is is what you have available to you today much different than just a couple of years ago?

[00:05:57]>> Oh, it's it's a revolution and it's incredibly exciting because we think not only can we be more efficient practitioners, but we can actually treat patients better. We can improve patient care because it will be much more personalized, it will be much more patient specific and we can really inform the patient in a discussion about what risk calculators and risk they're willing to take or their journey using these types of technologies.

[00:06:27]>> Yeah, do you think do you think people understand the the technology that is available today to actually make them feel better? I mean pain, you know, is is is something that everyone's feeling in terms of orthopedic at some point, right? I mean your your joints age.

[00:06:43]>> They can wear out.

[00:06:44]>> Yeah, they can wear out.

[00:06:45]>> Yeah, I mean I think these tools have to be created in such a way that they make them accessible for both patients and doctors. It's a huge amount of data.

[00:06:53]Synthesizing this data and making it presentable is really the trick.

[00:06:57]>> I love it. Dr. Pearl, great to have you this morning. Thanks very much for walking us through it.

[00:07:01]>> Thank you, Maria.

[00:07:02]>> Dr. Andrew Pearl from HSS, Hospital for Special Surgery. Don't miss AI week all week long here on Mornings with Maria.

[00:07:09]We will continue our focus on healthcare and defense this morning. Later on in the week we're taking a look at education and the impact of AI there.