NEW ORLEANS—An artificial intelligence model trained to analyze electrocardiogram (ECG) images identified which patients had transthyretin-mediated amyloid cardiomyopathy (ATTR-CM), according to findings scheduled for presentation at the upcoming American College of Cardiology 2026 Annual Scientific Session.
“Such a tool may improve ATTR-CM detection and help prioritize confirmatory testing in underserved communities,” wrote first author Philip Merio Croon, MD, an associate research scientist at the Yale School of Medicine, New Haven, Connecticut, and coauthors of the study abstract.
Researchers used data and ECG scans for 11,291 patients, of whom 293 had ATTR-CM, to develop the model, named ECGi-ATTR. They validated its performance in a separate cohort that included 65 patients with ATTR-CM and 20,157 control subjects.