An AI platform designed to identify various acute neurological illnesses, such as stroke, hemorrhage and hydrocephalus, recognized disease in CT scans in just 1.2 seconds, according to New York City-based Icahn School of Medicine at Mount Sinai.
Researchers published their findings in Nature Medicine.
Here are five things to know:
1. Researchers from Mount Sinai AI Consortium used 37,256 head CT scans to train a deep neural network to identify if an image's contents were critical or noncritical.
2. The platform was tested in a blinded, randomized controlled trial in a simulated clinical environment to assess and categorize head CT scans based on severity.
3. Compared to the amount of time it took a radiologist to notice a disease, the average time for the computer algorithm to preprocess an image, run its inference method and raise an alarm was 150 times shorter.
4. Senior author of the study Eric Oermann, MD, said the next phase of research will involve enhanced computer labeling of CT scans and a shift to "strongly supervised learning approaches" as well as novel techniques for increasing data efficiency.
5. Researchers estimate it will take up to two years to implement the changes to re-engineer the system.
"The expression 'time is brain' signifies that rapid response is critical in the treatment of acute neurological illnesses, so any tools that decrease time to diagnosis may lead to improved patient outcomes," study co-author Joshua Bederson, MD, concluded.