FDA-approved AI echocardiogram software bests cardiologists in reducing LVEF variability
A deep-learning software that can automatically calculate left ventricular ejection fraction (LVEF) with less variability than a cardiologist recently received approval June 19 from the U.S. Food and Drug Administration (FDA).
The fully automated algorithm—EchoMD AutoEF technology—estimated LVEF with a variability of 8.92 percent, which was slightly below the 9.2 percent average for cardiologists. This study was presented June 25 at the 2018 American Society of Echocardiography (ASE) Annual Scientific Sessions.
Created by Bay Labs in San Francisco, the software analyzes digital video clips from echocardiograms and chooses the highest quality clips to determine LVEF. The EcohMD AutoEF technology was trained on a curated dataset of more than four million images taken from 9,000 patients. It can be integrated into any DICOM PACS, according to a release.
The study included 405 echocardiograms from patients at the Minneapolis Heart Institute (MHI). The cohort represented a wide range of body mass index, EF values and ultrasound systems, according to the release.
“Historically there have been challenges with variability and reproducibility in reporting of the ejection fraction, especially when the EF is not normal; our study showed that the EchoMD AutoEF algorithms can aid interpretation enormously and have less variability than cardiologists reported in literature,” said co-author Richard Bae, MD, director of the echocardiography laboratory at MHI. “By supporting fast, efficient and accurate AI-assisted echocardiogram analysis, the algorithms can allow physicians to focus on putting results into context for the patient—guiding prognosis and course of management.”
Matt joined Chicago’s TriMed team in 2018 covering all areas of health imaging after two years reporting on the hospital field. He holds a bachelor’s in English from UIC, and enjoys a good cup of coffee and an interesting documentary.