Cedars-Sinai Researchers Use AI to Detect Liver Disease in Heart Scans 

(Antonio Marca/Shutterstock)

Liver disease affects 4.5 million people in the U.S. according to CDC estimates, and it can often be asymptomatic. Researchers at Cedars-Sinai are working to make diagnosis of liver disease easier and faster using an artificial intelligence algorithm.

Cedars-Sinai investigators have developed a machine learning model that can identify chronic liver disease from videos taken during an echocardiogram, which is a common test for heart disease. The test uses ultrasound to screen a patient’s heart for cardiovascular disease, and a standard echocardiogram study often contains more than 50 videos, including images of the liver.

The machine learning model is called EchoNet-Liver, and the researchers describe it in their corresponding study as a deep-learning, computer-vision pipeline that can identify high-quality subcostal images from full echocardiogram studies and detect the presence of cirrhosis and steatotic liver disease. EchoNet-Liver was developed using over 1.5 million echocardiogram videos from over 66,000 studies involving nearly 25,000 patients at Cedars–Sinai Medical Center.

“People with heart disease often develop chronic liver disease, and distinguishing between primary liver disease and liver injury secondary to heart disease can be challenging,” said David Ouyang, MD, a cardiologist in the Department of Cardiology in the Smidt Heart Institute, an investigator in the Division of Artificial Intelligence in Medicine, and a senior author of the study published in NEJM AI. “Our deep-learning model can help doctors spot liver disease that might have gone unnoticed and thus direct appropriate follow-up testing.”

The technology builds upon EchoNet, a computer-vision technology developed by Ouyang and colleagues that can identify and analyze patterns in echocardiograms, according to a Cedars-Sinai release.

David Ouyang, MD. (Source: Cedars-Sinai)

The study concludes that deep-learning assessment of echocardiograms enables opportunistic screening for steatotic liver disease and cirrhosis, helping to identify patients who may benefit from further diagnostic testing and treatment for chronic liver disease.

“Incorporating AI into echocardiograms, which capture images of the heart and the liver, can lead to a diagnosis of liver disease without additional costs,” said Alan Kwan, MD, assistant professor in the Department of Cardiology in the Smidt Heart Institute at Cedars-Sinai, and senior and corresponding author of the study.

Machine learning models like EchoNet-Liver highlight the growing potential of AI-driven computer vision in medical diagnostics. By leveraging existing imaging techniques, these models can enhance disease detection without adding cost or complexity to routine screenings. As AI continues to evolve, its ability to identify patterns in medical imaging could lead to earlier diagnoses, more targeted treatments, and improved patient outcomes across a range of conditions. The success of EchoNet-Liver shows how AI is currently transforming healthcare, offering new tools to detect diseases that might otherwise go unnoticed.