New AI Tool Helps Detect Hard-to-Diagnose Heart Disease

A new artificial intelligence (AI) tool created by physician-scientists in the Smidt Heart Institute at Cedars-Sinai was able to detect two types of heart disease that are typically hard to diagnose, according to a study published in JAMA Cardiology

Though there is a high prevalence of heart disease in America, several heart conditions often go unrecognized by healthcare professionals. Two of these conditions are hypertrophic cardiomyopathy and cardiac amyloidosis.

“These two heart conditions are challenging for even expert cardiologists to accurately identify, and so patients often go on for years to decades before receiving a correct diagnosis,” said David Ouyang, MD, a cardiologist in the Smidt Heart Institute and senior author of the study, in the press release.

To combat this issue and take steps to enhance cardiac care, physicians at the Smidt Heart Institute created an algorithm, which identifies specific features such as the thickness of heart walls and the size of heart chambers, to detect heart disease. The two-step algorithm was tested on over 34,000 cardiac ultrasound videos.

Before creating the AI tool, cardiologists had difficulty recognizing physical changes in the heart and defining whether they were related to disease or simply just to aging.

“The algorithm identified high-risk patients with more accuracy than the well-trained eye of a clinical expert,” said Ouyang. “This is because the algorithm picks up subtle cues on ultrasound videos that distinguish between heart conditions that can often look very similar to more benign conditions, as well as to each other, on initial review.”

Both hypertrophic cardiomyopathy and cardiac amyloidosis are diseases involving abnormal heart activity.

Hypertrophic cardiomyopathy is caused by heart muscle thickening, causing odd rhythms. Cardiac amyloidosis occurs because of amyloid, an unhealthy protein, building up within heart tissue. Although different, it is difficult distinguishing between these two diseases using traditional echocardiograms.

In addition to the recognition and differentiation of heart diseases, another potential benefit of the new algorithm is faster diagnosis. A quicker diagnosis can lead to quicker access to care and quicker recovery, reducing rates of mortality, hospitalizations, and heart failure.

Moving forward, clinical trials involving this AI algorithm will be launched. These trials will involve potential cardiac amyloidosis-diagnosis patients interacting with the Smidt Heart Institute physicians.

Along with the enhancement of care, there have been numerous instances when AI has provided physicians with reliable services. A study explains how using AI to administer drugs allows anesthesiologists to focus on more important tasks, such as ensuring a patient’s comfortability, eliminating pain, and retaining physiological stability.