AI & MD: What's Next?

Photo by  h heyerlein  on  Unsplash

Photo by h heyerlein on Unsplash

Post By: Molly Esselstrom, Upstream Research™ Marketing Manager

A recent New Yorker piece struck me as particularly relevant amidst current news that is digging deep into and questioning the consequences of Artificial Intelligence (AI). The piece opens on a woman in the hospital for symptoms of a stroke and the author’s musings about diagnostics: Could machines could learn to do it as well as humans? And, if so, what would the consequences be?

Interestingly, the author’s own introduction with diagnostics offers a look into the powers of a doctor’s deduction, of a real human with years of experience and studies under his or her belt making rational deductions based on a patient’s symptoms and behaviors. He brings up an important distinction in knowledge between “knowing that” (factual, propositional) and “knowing how” (implicit, experiential, skill-based).

“Early efforts to automate diagnosis tended to hew closely to the textbook realm of explicit knowledge.” This, in effect, eliminates the human aspect of diagnostics – the knowing how – and only reveals one part of a complete diagnosis. The balance would be the marriage of knowing that and knowing how with AI and automated diagnoses. In other words, algorithms that had “deep learning” capabilities.

The piece goes on to profile a computer scientist, Sebastian Thrun, who “envisages a world in which we’re constantly under diagnostic surveillance. Our cell phones would analyze shifting speech patterns to diagnose Alzheimer’s. A steering wheel would pick up incipient Parkinson’s through small hesitations and tremors. A bathtub would perform sequential scans as you bathe, via harmless ultrasound or magnetic resonance, to determine whether there’s a new mass in an ovary that requires investigation. Big Data would watch, record and evaluate you: we would shuttle from the grasp of one algorithm to the next.”

An interesting thought that arises from this: “many cancers are destined to be self-limited. We die with them, not of them. What if such an immersive diagnostic engine led to millions of unnecessary biopsies?” A good question and one that rides in the balance.

Many of the interviewed clinicians in the story had similar responses to AI; they welcomed the change if it helped them diagnose with greater accuracy and improved their patient’s lives. Key word: helped.

In working together, AI and MDs can create a more complete picture of health.


Upstream Research™ original post