Machine learning detects high-risk lung nodules in radiology reports

An automated artificial intelligence system can identify potentially cancerous lung nodules based solely on analysis of written radiology reports, according to new research published in the Journal of Clinical Oncology by Franklin, Tenn.-based AI company Digital Reasoning.

In the study, a system using machine learning and natural language processing analyzed nearly 9,000 narrative CT radiology reports. The technology was able to detect and triage high-risk pulmonary nodules in the text of the reports with over 90 percent precision.

"This approach promises to improve healthcare quality by increasing the rate of appropriate lung nodule incidental finding follow-up and treatment without excessive labor or risking overutilization," the study's authors wrote.

More articles about AI:
Optum funnels $16M into AI startup Health[at]Scale
IBM's AI distinguishes between healthy, cancerous cells in breast tumors
UPenn Medicine's AI tool for data analytics is open-source, free to the public

© Copyright ASC COMMUNICATIONS 2019. Interested in LINKING to or REPRINTING this content? View our policies by clicking here.

 

Top 40 Articles from the Past 6 Months