acm-header
Sign In

Communications of the ACM

ACM TechNews

Computer Detects Abuse Before Doctors


View as: Print Mobile App Share:

Ben Reis at Children's Hospital Boston and Harvard Medical School led a research team that developed software that could spot warning signs of domestic abuse. Reis and fellow researchers Isaac Kohane and Kenneth Mandl collected the medical records of 561,000 patients over six years in an unidentified U.S. state. Of that number, about 19,000 had known cases of domestic abuse. The program sorted through about two thirds of the files to spot common patterns of domestic abuse.

Although some of the factors the computer identified were already known by medical professionals, others were not. New symptoms include alcoholism in the case of women and depression and mood disorders in the case of men. Sometimes the software caught domestic abuse cases six years earlier than doctors, and other times not at all. On average, the computer caught domestic abuse cases one and a half years earlier than doctors.

"Sometimes doctors are blinded because they don't have access to 10 years of medical history," Reis says. "We're trying to empower the doctors to make the decision."

From New Scientist
View Full Article

 

Abstracts Copyright © 2009 Information Inc., Bethesda, Maryland, USA


 

No entries found

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