Publications

This study evaluated whether artificial intelligence applied to 12-lead ECGs (ECG-AI) can improve prediction of heart failure (HF) compared with traditional clinical models. Using pooled data from over 14,000 participants in major cohort studies, ECG-AI models detecting systolic and diastolic dysfun… Read More

JACC Journals 2025

November 12, 2025

This study demonstrates that AI applied to routine ECGs can identify asymptomatic left ventricular dysfunction (ALVD) with high accuracy, achieving an AUC of 0.93, sensitivity of 86.3%, and specificity of 85.7%. In patients without current dysfunction, a positive AI screen indicated a fourfold incre… Read More

Mayo Clinic Proceedings: Digital Health

October 25, 2024

This study assessed the effects of gender-affirming hormone therapy (GAHT) on ECG patterns in transgender individuals using an AI algorithm. Among transgender women (TGW), GAHT significantly lowered the probability of a male ECG pattern, while among transgender men (TGM), it increased the probabilit… Read More

European Heart Journal

October 14, 2024

This study evaluated echocardiographic characteristics and mortality risk in patients with false positive (FP) results from an AI-based ECG model that detects low ejection fraction (EF). FP patients had more echocardiographic abnormalities than true negatives (TN), and 97% of them showed some abnorm… Read More

ScienceDirect

September 27, 2024

This study in Nigeria investigated AI-guided screening for diagnosing left ventricular systolic dysfunction (LVSD) in pregnant and postpartum women. Participants were randomized to either AI screening using digital stethoscopes and ECGs or usual care, with 3.4% in the AI group and 2.0% in the contro… Read More

Nature Medicine

September 1, 2024