AI-Enhanced Ultrasound Enables Early Detection of Cardiac Amyloidosis
A new artificial intelligence (AI) model developed through collaboration between Mayo Clinic and Ultromics, Ltd., an AI echocardiography company based in Oxford, England, demonstrates high accuracy in screening for cardiac amyloidosis. This rare and progressive form of heart failure has historically been difficult to diagnose early, but this AI tool shows promise in changing that.
Accurate Screening with a Single Echocardiography Videoclip
The AI model, the first of its kind, was tested and validated on a large, multiethnic patient population, including input from researchers at the University of Chicago Medicine and international collaborators. Published in the European Heart Journal, the study highlights the model’s performance with 85% sensitivity and 93% specificity. This means the AI correctly identifies 85% of patients with cardiac amyloidosis and 93% of those without the disease.
Importantly, the AI uses a standard echocardiography videoclip and can differentiate cardiac amyloidosis from other heart conditions with similar presentations. It performs well across all major types of cardiac amyloidosis, offering a practical diagnostic advantage.
Why Early Detection Matters
Cardiac amyloidosis occurs when abnormal protein deposits, called amyloid, build up in the heart. This causes the heart muscle to stiffen, impairing its function and leading to heart failure. Because symptoms and imaging results can mimic other cardiac diseases, early diagnosis is often missed.
Early detection is critical because new drug therapies can slow or stop disease progression when started promptly. However, diagnosing cardiac amyloidosis traditionally requires extensive and burdensome testing.
“Cardiac amyloidosis can be challenging to diagnose because it’s often difficult to distinguish from other heart issues without a burdensome amount of testing,” said a co-lead researcher from the University of Chicago Medicine.
Building on Previous Advances in AI Echocardiography
This AI model builds on prior work by Mayo Clinic and Ultromics, who developed an AI echocardiography tool to detect heart failure with preserved ejection fraction (HFpEF). That model received FDA clearance in 2022. HFpEF affects many patients and is difficult to diagnose; about 15% of HFpEF patients have underlying cardiac amyloidosis.
Clinical Impact and Integration
The amyloid detection AI model is FDA-cleared and already in use at several U.S. medical centers. Clinicians report that the model outperforms traditional clinical and echocardiographic screening methods, providing clearer guidance on which patients require further confirmatory tests.
One cardiologist involved in the study noted, “This AI model helps identify patients earlier so they can receive the treatment they need.” The model’s ability to analyze a common echocardiography view means it can integrate smoothly into daily clinical workflows without adding complexity or compromising accuracy.
Support and Future Use
The study was partially funded by Ultromics, and Mayo Clinic holds financial interests in this technology, with revenues supporting its non-profit mission in patient care, education, and research.
As this AI tool becomes more widely adopted, it offers healthcare professionals a practical approach to detect cardiac amyloidosis earlier, enabling timely intervention and improving patient outcomes.
Your membership also unlocks:
