UPMC introduces new A.I. smartphone app to help doctors diagnose ear infections

PITTSBURGH — Doctors at the University of Pittsburgh Medical Center have been developing a smartphone app for more than a decade that uses artificial intelligence to identify features that are highly predictive of the way doctors think when diagnosing an ear infection.

“It’s an area where there’s a lot of misdiagnosis,” said Dr. Alejandro Hoberman, director of UPMC General Academic Pediatrics.

He’s talking about ear infections in children, telling Channel 11 there’s evidence that somewhere between 12-30% of prescriptions are unnecessarily written.

“In order to get better at diagnosing, it needs to get better at predicting the features that lead to that diagnosis,” Dr. Hoberman said.

That was a driving force behind this new technology that uses A.I. to analyze pictures and videos of an eardrum to see if abnormal cells can be identified. He’s able to capture and classify videos using an oroscope attached to a smartphone.

“We send it to the Cloud, where it compares with 1,151 videos that we have annotated with the findings of experts evaluating those videos,” Dr. Hoberman said.

Dr. Hoberman said there are different types of ear infections, but not all of them need to be treated with antibiotics, which makes technology like this a game changer.

“This will enhance the diagnosis…make it sharper and allow it that the kids who actually need the treatment will receive the treatment,” Dr. Hoberman said.

Right now, the technology is being used in clinical research, at several UPMC pediatric offices, and for teaching purposes. Eventually, he hopes it can be used for telemedicine purposes and in other offices as well.

Download the FREE WPXI News app for breaking news alerts.

Follow Channel 11 News on Facebook and Twitter. | Watch WPXI NOW