Multimedia

Watch: Using AI to Diagnose Disease

05 February 2018

An ATM can read handwriting on a check. Facebook can suggest the correct name of the person to tag in a picture. And not long from now, computer vision powered by artificial intelligence will be able to interpret medical scans and diagnose conditions.

A first step in developing that technology has been reported in the journal Radiology, in which researchers describe using machine learning techniques, including natural language-processing algorithms, to identify clinical concepts in radiologist reports for CT scans.

In order for computer vision to be effective in medicine, a machine must be able to tell the difference between normal and abnormal findings. For the study, conducted at Icahn School of Medicine at Mount Sinai, researchers used over 96,000 radiologist reports associated with head CT scans to train computer software to understand clusters of phrases.

"The language used in radiology has a natural structure, which makes it amenable to machine learning," says senior author Dr. Eric Oermann, M.D., an instructor in the Department of Neurosurgery at the Icahn School. "Machine learning models built upon massive radiological text datasets can facilitate the training of future artificial intelligence-based systems for analyzing radiological images."

The techniques used in the study yielded an accuracy of 91 percent, demonstrating that it is possible to automatically identify concepts in text from the complex domain of radiology.

"The ultimate goal is to create algorithms that help doctors accurately diagnose patients," says medical student John Zech, who served as first author for the research. "Deep learning has many potential applications in radiology — triaging to identify studies that require immediate evaluation, flagging abnormal parts of cross-sectional imaging for further review, characterizing masses concerning for malignancy — and those applications will require many labeled training examples."

Dr. Joshua Bederson, study co-author, professor and system chair for the Department of Neurosurgery, calls this kind of research the “critical first step in harnessing the power of artificial intelligence to help patients."

To contact the author of this article, email tony.pallone@ieeeglobalspec.com


Powered by CR4, the Engineering Community

Discussion – 0 comments

By posting a comment you confirm that you have read and accept our Posting Rules and Terms of Use.
Engineering Newsletter Signup
Get the Engineering360
Stay up to date on:
Our flagship newsletter covers all the technologies engineers need for new product development across disciplines and industries.
Advertisement
Advertisement

Upcoming Events

Advertisement