Evaluation of Artificial Intelligence Software US2.AI for Automated Echocardiographic Measurements Compared to Expert Analysis in Adults Undergoing Transthoracic Echocardiography

What this Trial Is About

This observational study focuses on adults undergoing standard transthoracic echocardiography to evaluate the diagnostic accuracy of US2.AI, an artificial intelligence software. The software is designed to automatically analyze two-dimensional and Doppler echocardiographic images from various ultrasound machines, aiming to measure cardiac structure and function accurately. Researchers compare these automated measurements with those taken by experienced echocardiographers to assess agreement and reliability. Participants receive routine echocardiographic exams performed by expert operators, with images analyzed both manually and automatically by the AI software. No additional imaging or follow-up visits are required beyond standard clinical practice. The study occurs across seven high-volume echocardiography centers, allowing the evaluation of the software in real-world clinical settings. During the study, measurements such as chamber dimensions, ventricular function, myocardial strain, and Doppler parameters are collected. Researchers assess the time to complete analyses and compare AI-derived results with expert assessments to determine accuracy. Results may support AI use to standardize echocardiography and assist in identifying cardiovascular conditions. The study is expected to complete in January 2027.

Diagnostic Accuracy of a Deep Learning-Based Software for Automated Multiparametric Echocardiographic Measurements