Innovative AI Heart Failure Diagnostic Tool Achieves 85% Accuracy Using Routine Ultrasound Data In Clinical Study

New AI model predicts advanced heart failure with 85% accuracy using routine ultrasound data, potentially saving thousands of cardiac patients annually.

By: AXL Media

Published: Mar 21, 2026, 6:57 AM EDT

Source: Information for this report was sourced from Weill Cornell Medicine

A Technological Breakthrough In Cardiac Diagnostics

The Cardiovascular AI Initiative, a collaborative effort involving Weill Cornell Medicine and NewYork-Presbyterian, has unveiled a machine learning model designed to streamline the identification of advanced heart failure. According to Dr. Fei Wang, the study's senior author, this innovation utilizes data sources already integrated into routine medical care to provide a more efficient assessment pathway. By processing moving ultrasound images and heart valve dynamics, the tool aims to bridge a critical gap in cardiovascular medicine where complex diagnostic requirements often prevent patients from receiving timely, life-saving interventions.

Overcoming The specialized Equipment Bottleneck

Traditionally, diagnosing the severity of heart failure relies on cardiopulmonary exercise testing, a process that requires expensive specialized equipment and a highly trained staff. This infrastructure is typically confined to major metropolitan medical centers, leaving an estimated 200,000 Americans with advanced heart failure struggling to access appropriate care. The new AI-powered approach seeks to decentralize this process by transforming standard echocardiograms, which are widely available in smaller clinics, into powerful predictive tools for assessing patient risk.

Collaborative Engineering Between Clinicians And Researchers

The development of this tool was driven by an intensive partnership between over 40 heart failure specialists and computer science experts from Cornell Tech and the Cornell Ann S. Bowers College of Computing and Information Science. Dr. Deborah Estrin noted that the close interaction between medical doctors and AI researchers led to the creation of entirely new machine learning techniques. This synergy allowed the team to build a multi-modal model that can interpret complex waveform imagery and blood flow data alongside traditional electronic health records.