A breakthrough artificial intelligence model that has a classifier for predicting tasks and a "rejector" to identify whether it or a human expert should complete tasks yielded 8% improved performance in diagnosing collapsed lung and enlarged heart in chest X-rays, compared with human readers or AI alone, according to findings published on arXiv.org. "We hope that our method will inspire machine learning practitioners to get more creative in integrating real-time human expertise into their algorithms," said researcher David Sontag.
Sign up for AdvaMed SmartBrief
Medical technology news
Get the intelligence you need: news and information that is changing your industry today, hand-curated by our professional editors from thousands of sources and delivered straight to your inbox.