How AI Can Enhance Diagnostic Tools for Cardiologists

San Francisco, CA, – UCSF researchers have developed an AI-based system to predict cardiac pumping performance, reducing the need for invasive testing and improving patient care. The system, called CathEF, utilizes deep neural networks to estimate left ventricular ejection fraction (LVEF) from standard angiogram videos. The researchers conducted a study with 4042 angiograms and corresponding transthoracic echocardiograms, demonstrating the accuracy of CathEF in predicting LVEF. The algorithm was externally validated and performed well across different patient demographics and clinical conditions. The findings offer new possibilities for noninvasive assessment and real-time information during angiography, ultimately enhancing clinical decision-making.

Authors: Dr. Geoff Tison and Dr. Robert Avram led the study, with additional authors from UCSF, including Joshua P. Barrios PhD, Sean Abreau MS, and Jeffrey E. Olgin MD.

Funding: The study was supported by US NIH grants K23HL135274 and U2CEB021881.

About UCSF Health: UCSF Health is a world-renowned academic medical center known for its innovative patient care, advanced technologies, and groundbreaking research. It includes UCSF Medical Center, UCSF Benioff Children’s Hospitals, Langley Porter Psychiatric Hospital and Clinics, UCSF Benioff Children’s Physicians, and the UCSF Faculty Practice. UCSF Health has affiliations with hospitals and health organizations throughout the Bay Area.

For more information, visit: https://www.ucsfhealth.org/

Follow UCSF: ucsf.edu | Facebook.com/ucsf | Twitter.com/ucsf | YouTube.com/ucsf