AI can diagnose acute illnesses faster than human

AI can diagnose acute illnesses faster than human

Scientists have developed an artificial intelligence platform that can identify the disease in brain CT scans in 1.2 seconds, and diagnose a range of acute neurological illnesses, such as stroke, and haemorrhage.

The study, published in the journal Nature Medicine, shows that the system was faster than human diagnosis.

"With a total processing and interpretation time of 1.2 seconds, such a triage system can alert physicians to a critical finding that may otherwise remain in a queue for minutes to hours," said Eric Oermann, at the Icahn School of Medicine at Mount Sinai in the US.

"We're executing on the vision to develop artificial intelligence in medicine that will solve clinical problems and improve patient care," said Oermann.

This is the first study to utilise artificial intelligence for detecting a wide range of acute neurologic events and to demonstrate a direct clinical application.

Researchers used 37,236 head CT scans to train a deep neural network to identify whether an image contained critical or non-critical findings.

The platform was then tested in a blinded, randomised controlled trial in a simulated clinical environment where it triaged head CT scans based on severity.

The computer software was tested for how quickly it could recognise and provide notification versus the 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, if necessary, raise an alarm was 150 times shorter than for physicians to read the image.

This study used "weakly supervised learning approaches," which built on the research team's expertise in natural language processing and the Mount Sinai Health System's large clinical datasets.

Oermann said the next phase of this research will entail enhanced computer labelling of CT scans and a shift to "strongly supervised learning approaches" and novel techniques for increasing data efficiency.

Researchers estimate the goal of re-engineering the system with these changes will be accomplished within the next two years.

"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," said Joshua Bederson, a professor at Mount Sinai Health System.

"The application of deep learning and computer vision techniques to radiological imaging is a clear imperative for 21st-century medical care," said Burton Drayer, from the Mount Sinai Health System. 

DH Newsletter Privacy Policy Get top news in your inbox daily