acm-header
Sign In

Communications of the ACM

ACM TechNews

AI Comes to the Operating Room


View as: Print Mobile App Share:
The dark ovals are tumor cells, among nerve fibers that appear as white streaks, indicating a malignant tumor called a diffuse glioma.

Surgeons are utilizing artificial intelligence and new imaging methods to diagnose brain tumors faster than pathologists, and with similar accuracy.

Credit: Michigan Medicine

Surgeons are utilizing artificial intelligence and new imaging methods to diagnose brain tumors faster than pathologists, and with similar accuracy.

Traditionally, sample tissue from the brain is sent to a lab for analysis through a microscope, which can take about 30 minutes. The new technique takes less than three minutes, using lasers to create images of the sample tissue after it has been removed from the brain and a computer to read the images in the operating room.

The researchers used images of tissue samples taken from 415 brain surgery patients to train an artificial intelligence system to identify the 10 most common types of brain tumor.

Overall, the system misdiagnosed 14 cases that doctors got right, while doctors missed 17 cases that the computer got right.

Neurosurgeon Daniel A. Orringer at NYU Langone Health said the study shows "the combination of an algorithm plus human intuition improves our ability to predict diagnosis."

From The New York Times
View Full Article - May Require Paid Subscription

 

Abstracts Copyright © 2020 SmithBucklin, Washington, DC, USA


 

No entries found

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