PX: Will AI Affect Medicine's Evolution?

Are physicians putting too much trust in artificial intelligence (AI)? How much has medicine evolved with AI, and what will medicine look like in the future? Our expert panel weighs in:

Leveraging Big Data and AI in the Hospital of Tomorrow

A Healthcare Analytics News™ Peer Exchange®

Segment 7/11

Kevin R. Campbell, M.D.: I’ve got a question for you, John, and this may be a setup but: is it possible to put too much trust in AI? And I know that’s a bit of a loaded question yet again.

John Nosta, B.A.: To me, AI is not a fixed reality but a journey. I think that to try to establish it in the context of the here and now is a false premise. What interests me is that we spend a lot of time worrying about the AI error. Does anybody spend time thinking about the physician error?

Kevin R. Campbell, M.D.: We do; we do every single day.

David E. Albert, M.D.: All the time.

John Nosta, B.A.: The horror of practice errors?

Kevin Campbell, M.D.: We have M&M [morbidity and mortality] rounds —

John Nosta, B.A.: Those are big problems

Kevin Campbell, M.D.: We teach our Fellows to do.

>> Watch the last episode: PX: Can AI Be Smarter Than Humans?

John Nosta, B.A.: But that’s the best we’ve got. The best we’ve got is the human construct. And I’m just saying that I think we could do better. Simple as that.

David E. Albert, M.D.: And I think we will. I mean we all are hopefully optimistic. I tell my children — I have four kids — and I tell them I want them to go do something and be happy. But I tell them there are two jobs that I really don’t want them to go into: truck driving and diagnostic radiology because AI will take those jobs away. It may not be this year, it may be in 25 years, but truck driving is going to go away because you go from point A to point B, and we know that it’ll be able to go from point A to point B. That robot truck will be able to do it.

The same thing with diagnostic radiology. Diagnostic radiology — looking at films. They’re going to not miss things, they’re going to do it better. And so you know that’s a little tongue in cheek but not really. And so, I think that it will come in stages. You’re right, it’s a journey, John, it’s an absolute journey, like the Gartner Hype Cycle is a cycle.

John Nosta, B.A.: And is that journey, if you draw the dots of artificial intelligence, linear or asymptotic? Does it become, at a point, where the threshold becomes a sense of humanity that it can get closer and closer and closer as it rises up, but it never passes the threshold? I think that is probably less clinical medicine and more philosophy, but I think that is where we are now, we are on that exponential journey. I think it would serve us well to be less arrogant as humans and think about the very nature of change in the world today.

David E. Albert: Well, we don’t have Skynet — machines creating machines — yet, but I think humans are in charge today. And I think where this goes is in our control, and I am very optimistic. I may not be quite with my friend, Vinod [Khosla], but I’m very optimistic that AI’s going to play an important role in addressing some of the problems we have in medicine: errors, costs, things of that nature. You know, access.

Kevin R. Campbell, M.D.: Prediction, prediction and intervention prior to disease.

David E. Albert, M.D.: As Geeta said, “Prediction is prevention.” If we can predict you’re going to have a fall, we can potentially prevent that fall. If we can predict you’re going to become septic in the hospital, we can prevent that sepsis and prevent that potential death, so I’m very optimistic. And I also believe we’re just at the beginning of this journey. Thirty years from now we’ll look back and this will be the infancy of AI in medicine and I just hope I’m around to see it.

John Nosta, B.A.: If we’re bold enough to look back and talk about the profound changes that have come in the past 30 years, keeping in mind that the cellphone is seven or eight-years-old, where does that put us 30 years from now? I think it’s extraordinary and it will tear at the fabric of humanity. It’s going to be a very, very interesting ride.

David E. Albert, M.D.: But remember, as has been stated before, the future’s already here, John, it’s just not evenly distributed. And that will remain the same, although, you can go to Africa, South America, you can go anywhere in the developing world and people have a super computer, globally connected in their pocket that cost them $50. And that revolutionary tool, which by the way is only 11-years-old, will be your primary care access point. That will be your banking access point. That will be your education access point — speaking as one who has two kids in private college.

John Nosta, B.A.: Not the physician? So the access point to care will be technology?

David E. Albert, M.D.: Because there aren’t enough physicians, John. There aren’t enough, we aren’t going to make more physicians. We built a bunch of new medical schools, but demand still outstrips supply. You know if you’re in a subspecialty like Geeta and my wife — unbelievably underserved.

Kevin R. Campbell, M.D.: Their job market is huge.

David E. Albert, M.D.: It takes months to get in to see my wife because, by the way, it’s not a fancy, procedure-driven specialty; it’s a cognitive specialty. And the number two thing is, everybody hurts, everybody’s joints hurt when they get older. I can verify that.

Geeta Nayyar, M.D., MBA: And that’s why also, telemedicine, virtual medicine, we talked earlier about Nicklaus Children’s [Hospital] — you know in South Florida they’re doing an enormous amount of work in telemedicine, and it’s really so important. I can tell you as a rheumatologist, pediatric rheumatology, you can’t find one.

David E. Albert: You can’t, no.

Geeta Nayyar, M.D., MBA: Guess who you go to? Me.

David E. Albert, M.D.: No, it goes to my wife. My wife went and did six months in Cincinnati at their children’s hospital so that she could see pediatric rheumatology patients, as an internist, and she was scared every time she’d go in with a baby who was having some kind of autoimmune problem. But this is the problem we’re not going to solve. The way we’re going to augment that is with technology, John, you’re right.

Kevin R. Campbell, M.D.: I would totally agree. I just got back from a mission trip during which I sailed all over the South Pacific, went to remote islands in Fiji, some places they had never seen a white person or doctor. Every person on that island had access to someone who had a cellphone. They’re all on Facebook. I saw a patient who had neurofibromatosis, and that’s not my area of expertise.

I took a picture. I uploaded it to a private discussion group on Facebook, and there were some folks there who had expertise in that area and they said, “Yes, you’re exactly right. This is what you need to do about this problem.” That is in a remote island where they have no running water, no electricity, they barely subsist on fishing and farming, yet they have this power and we’re able to bring care to them, and that’s what this is all about.

Thank you all for this very spirited discussion. This has been a great segment and I look forward to continuing on.

Get the best executive insights directly to your inbox. Sign up for our daily newsletter.

Related

PX: Is AI a Physician's Tool or Future Physician?

Machine Learning: The Future of Medicine?

How AI Is Shaking Up Healthcare, Beyond Diagnostics