AI‑powered ultrasound system unveiled in Turin aims to cut unnecessary C‑sections

A new intrapartum ultrasound system integrating artificial intelligence was presented at the Aiic clinical engineering conference in Turin. The AI analyzes trans‑perineal ultrasound images in real time and assigns a colour‑coded risk level – red, yellow or green – to indicate the likelihood of a successful operative vaginal delivery. In a validation study of 180 scans, the system achieved up to 97 % discriminative accuracy for the high‑risk (red) class and reported no cases where a safe delivery was misclassified as high risk. Developed by a team that includes researchers from CNR‑IFC, the University of Parma, the hospital in Lecce and the Amolab spin‑off, the tool is designed to replace the subjective digital vaginal examination, reduce unnecessary cesarean sections, improve maternal and neonatal safety, and lower healthcare costs. The Delivery Color Code aims to provide clinicians with an objective, standardized assessment to guide decision‑making during labour.