acm-header
Sign In

Communications of the ACM

ACM TechNews

How Doctors Use AI to Help Diagnose Patients


View as: Print Mobile App Share:

Physicians say AI holds promise, but they are also wary of yielding to machines, both because the technology is still emerging and because research has shown that biases in AI can harm the care some people receive.

Credit: Sam Kelly/The Wall Street Journal, iStock

Artificial intelligence (AI) is enhancing physicians' diagnostic abilities to better identify patients who might be vulnerable to certain ailments.

Medical AI programs employ an algorithm or set of algorithms that can determine patients' risk of health issues based on data, typically from electronic health records.

Paul Friedman at Mayo Clinic's cardiology department said doctors use an AI tool to spot heart failure and atrial fibrillation from electrocardiograms, potentially years before they might otherwise be diagnosed.

Last year, the Cano Health physician group used AI to identify diabetic retinopathy within images from an eye camera.

Suchi Saria at health AI company Bayesian Health said doctors in Baltimore can algorithmically identify septic hospitalized patients; research found patients could be treated nearly two hours earlier on average with AI diagnosis, reducing sepsis' hospital mortality rate by 18%.

From The Wall Street Journal
View Full Article - May Require Paid Subscription

 

Abstracts Copyright © 2023 SmithBucklin, Washington, DC, USA


 

No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account