Curie Brief
Turn on cookies to sign in
Signing in saves your progress to your Curie account. We can only do that with cookies on — turn them on to continue.
News
AI could help doctors catch what they might miss. Ophthalmologist Dr. John Hovanesian argues that medical AI should go beyond ambient scribing and help flag rare but serious diagnoses — like measles masquerading as pink eye — especially when physicians are pressed for time. The key? Giving AI a longitudinal view of patient history while keeping doctors firmly in the driver's seat.
AI in medicine: Beyond scribing to smarter diagnosis
Medical AI is mostly being used today as a glorified note-taker — and ophthalmologist Dr. John Hovanesian thinks we're selling it short. In a new opinion piece, he makes the case that AI's real value lies in helping time-pressed physicians catch rare but serious diagnoses they might otherwise miss. His example: a child with red eyes, runny nose, and a cough who gets labeled as viral conjunctivitis, when the real culprit could be measles — a disease with specific treatments and the potential for permanent eye and neurological damage.
Hovanesian isn't calling for AI to replace physician judgment. Instead, he envisions AI as a "colleague in the background" — one that quietly flags unusual possibilities and generates differential diagnoses, particularly when something doesn't quite add up. To do this well, AI needs access to a patient's full longitudinal history, which means the medical community will need to get more comfortable with privacy, security, and medicolegal concerns.
Key Takeaways
Why it matters: As physician burnout and time constraints grow, AI that nudges doctors toward overlooked diagnoses — rather than just transcribing visits — could meaningfully reduce diagnostic errors and improve patient outcomes.
