Learn about this first
Google said on X that its prenatal model can limit gestational age error to within 4 days in a prospective clinical study, and that it considers this level of accuracy sufficient to produce meaningful clinical impact. Google also said that expanding it to resource-limited regions could have a substantial effect on maternal mortality. These accuracy and impact claims are Google’s own statements.
Google Research subsequently published a prospective multicenter study in JAMA Network Open evaluating the generalizability of an AI gestational-age assessment model based on “blind-scan” ultrasound across different clinical settings and devices, stating that its accuracy was comparable to that of expert operators. The study is a peer-reviewed paper, but reports did not specify whether it was the same study as the one associated with the 4-day accuracy mentioned in the X post.
AI-generated from reports · updated 26 days agoLatest developments
Google Research publishes a study on AI gestational age estimation from blind-sweep ultrasound, with accuracy comparable to expert scans
Event progress2
Google Research Publishes Study on AI-Assisted Ultrasound Gestational Age Assessment
Google Research: Blog: Google Research Publishes Study on AI Blind-Scan Ultrasound Gestational Age Assessment, with Accuracy Comparable to Expert OperatorsNew Developments in Google’s Prenatal Model Clinical Study
Google: New Developments in Google’s Prenatal Model Clinical Study