Physicians versus computers – the wrong question!!!


Category : Medical Rants

Over the past 2 days, listening to separate podcasts, I have heard the same story and now have a better understanding of artificial intelligence.  A Freakonomics podcast – The Future (Probably) Isn’t as Scary as You Think

And in general, what’s happened in the past couple of years is the best chess player on this planet is not an AI. And it’s not a human. It’s the team that he calls centaurs; it’s the team of humans and AI. Because they’re complementary. Because AIs think differently than humans. And the same of the world’s best medical diagnostician is not Watson, it’s not a human doctor. It’s the team of Watson plus doctor.

In this Knowledge Project podcast, the same examples and conclusions develop – Pedro Domingos on Artificial Intelligence

As I listened to these discussions, an important concept crystallized.  Computers and especially AI will provide us the best “peripheral brain”, but they really cannot replace us.  Humans have strengths that computers likely will have great difficulty duplicating.  Computers do not really think, but their algorithms can reveal connections that we humans do not recognize.  So working with algorithms we can do a better job at diagnosis and treatment.

Recently we had a wonderful gentleman in his 70s admitted for chest pain.  Chest pain occurs for many different reasons.  Our human skills are put to a major test in sorting out a precise history.  This intelligent, verbal patient kept talking about his heartburn.  He pointed to his sternum and described the discomfort as burning.  He had a difficult time defining relieving or inducing activities.  He had no major risk factors for CAD other than his gender and age.

While talking the history, we asked him if he still worked.  He told us that he had to stop working as a janitor because of the chest discomfort.  He further stated that he had slowed down at first, because his regular pace brought on the pain.

We got this history despite having asked him previously what brought on the pain.  A different approach clarified the pain in our minds.

His nuclear medicine stress test showed an approximately 10-15% perfusion/reperfusion defect in the right coronary artery region.

We called cardiology, who took the history and believed that the patient really had reflux.

One week later when reflux medications and CAD treatment (including beta blocker, statin and ASA) did not relieve the pain, and it actually worsened, a cardiac catheterization showed an isolated 95% right coronary artery lesion.  A stent totally relieved his discomfort (which was even occurring at rest).

AI would only work here if it received a proper history.  But taking the history required some skill and some intuition.  Human diagnosis would only work if we paid attention to the technology.  We need to combine an atypical history with other findings.  AI would work if a human could interpret a history that was not linear in any way.

So the real question is how do we partner with artificial intelligence?  We already partner with Dr. Google and Dr. Uptodate.  Partnering with AI in selected cases can help us do our jobs better.  Together our patients will benefit.

Comments (4)

This is not news to some of us.
When CTs were introduced, there was a huge debate – what was better a CT or a good neurologist? Both sides mounted their arguments, The answer didn’t take long – the answer was a good neurologist using a CT.
In my area it was the same with PA catheters – but the majority of the critical care community never caught on to that so PA catheters fell out of favor.
A combination fo a good physician and AI will always be better than either by itself.
But, and this si important, it will be a good doctor using AI not th either way around.

There are more things in heaven and earth than are dreamed of in your algorithms. And even Spock had to take a guess.

Chest pain. Abnormal stress test. “Heartburn”. Cardiology defers angio? I keep reading that part over and over because I find it not really believable. Maybe we should say Board Certified practicing Internist> AI and cardiologist

I agree with Frank L. I had to read the post three times since the vignette did not make sense–man in his seventies with substernal pain described as heartburn severe enough to stop working, and stress test showing myocardial ischemia, and the cardiologist diagnoses GERD and does not arrange for angiogram? Was this on Jul 1st?

Post a comment