The study, published online Monday in the journal BMJ Quality and Safety, underscores the trade-offs in the diagnostic process: More medical tests generally improve detection, but they can also increase costs, harm patients, and produce false positives.

WSJ: Adding Up Diagnosis Errors

The company claims that its learning algorithms can be used to process data for a wide variety of applications, from predicting the effectiveness of sales promotions to fraud detection to molecular modeling to astronomy.

FORBES: Skytree Uses Machine Learning To Crunch Big Data