New York City-based Icahn School of Medicine at Mount Sinai researchers have developed an artificial intelligence system that links unconnected medical events over time to improve diagnostic accuracy.
The system, called Inference on Electronic Health Records, or InfEHR, connects scattered data in EHRs to reveal patterns that may indicate hidden diseases, according to an Oct. 15 news release from the health system. It was designed by Mount Sinai’s Windreich Department of Artificial Intelligence and Human Health in collaboration with other institutions.
In a study published Sept. 26 in Nature Communications, InfEHR analyzed deidentified records from Mount Sinai and UC Irvine hospitals. The AI tool identified neonatal sepsis 12 to 16 times more accurately and postoperative kidney injury 4 to 7 times more accurately than current methods.
Mount Sinai researchers plan to make InfEHR’s code available to other institutions to further explore its use in personalized treatment and research applications, according to the study.
