acm-header
Sign In

Communications of the ACM

ACM TechNews

Regenstrief, Iu Study Finds Machine Learning as Good as Humans' in Cancer Surveillance


View as: Print Mobile App Share:
Senior study author Shaun Grannis.

Senior study author Shaun Grannis, interim director of the Regenstrief Center of Biomedical Informatics, says the new finding constitutes "a major infrastructure advance."

Credit: IUSM Newsroom

Researchers at Indiana University (IU) and the Regenstrief Institute have found existing algorithms and open source machine-learning tools are as good as, or better than, human reviewers in detecting cancer cases using data from free-text pathology reports.

The researchers sampled 7,000 free-text pathology reports from more than 30 hospitals participating in the Indiana Health Information Exchange. The researchers used open source tools, classification algorithms, and varying feature selection approaches to predict if a report was positive or negative for cancer. The study found a fully automated review produced results similar to or better than those of trained human reviews.

The results also demonstrate machine learning can greatly facilitate the process of reporting cancer cases. "We think that it's no longer necessary for humans to spend time reviewing text reports to determine if cancer is present or not," says senior study author Shaun Grannis, interim director of the Regenstrief Center of Biomedical Informatics.

"This is not an advance in ideas, it's a major infrastructure advance--we have the technology, we have the data, we have the software from which we saw accurate, rapid review of vast amounts of data without human oversight or supervision," Grannis says.

From IUSM Newsroom
View Full Article

 

Abstracts Copyright © 2016 Information Inc., Bethesda, Maryland, USA


 

No entries found