Cardiology
21 November, 2026
J Am Coll Cardiol. 2026 Nov 7:S0735-1097(26)10299-X. doi: 10.1016/j.jacc.2026.10.066. Online ahead of print.
ABSTRACT
BACKGROUND: Excess epicardial adipose tissue (EAT) is associated with cardiovascular-kidney-metabolic dysfunction, but its assessment has traditionally required advanced imaging.
OBJECTIVES: We investigated whether artificial intelligence (AI)-enabled echocardiography could characterize epicardial adiposity and identify residual cardiometabolic risk.
METHODS: We developed PanAdipo, a video-based deep learning model in 1,114,441 videos from 28,797 studies across the Yale-New Haven Health System (YNHHS) (2016-2022), using report-derived expert reader labels of prominent EAT (2.3% of studies). PanAdipo was evaluated in 5 testing cohorts: a temporally distinct YNHHS transthoracic echocardiography set (n = 4,588 patients contributing 5,130 studies), an emergency department point-of-care ultrasound cohort (n = 10,957), the geographically distinct MIMIC-IV (Medical Information Mart for Intensive Care, version IV) data set (n = 4,549), the community-based MESA (Multi-Ethnic Study of Atherosclerosis) Exam 6 (n = 2,740), as well as in held-out YNHHS patients with paired coronary computed tomography (CT) angiography within 12 months (n = 5,167). Analyses examined discrimination of the report label and of CT-defined high indexed EAT volume, correlation with paired cardiac CT phenotypes, and body mass index (BMI)-adjusted associations with cardiometabolic biomarkers and newly documented metabolic disease.
RESULTS: In the held-out health system test set, PanAdipo identified prominent EAT documented in clinical reports with an area under the receiver operating characteristic curve of 0.86 (95% CI: 0.82-0.91) across views. Against paired CT-derived measurements, PanAdipo discriminated age- and sex-specific high EAT (EAT / heart volume ≥95th percentile) with an area under the receiver operating characteristic curve of 0.90 (95% CI: 0.89-0.92) overall and of 0.91-0.95 within age-sex groups, and correlated more strongly with indexed EAT volume (Spearman ρ = 0.76) than with BMI (Spearman ρ = 0.41; all P < 0.001); PanAdipo alone explained 53% of indexed EAT volume variance vs 23% for age, sex, and BMI combined. In MESA, higher scores were associated with an adverse metabolic profile (higher homeostatic model assessment for insulin resistance [β = 0.17 per SD; 95% CI: 0.11-0.22; n = 950]) after adjustment for age, sex, and BMI. In exploratory health system cohort analyses, higher scores were associated with newly documented steatotic liver disease (pooled BMI-adjusted HR: 1.33 [95% CI: 1.11-1.59] per SD of log[PanAdipo]) and hyperlipidemia (HR: 1.11; 95% CI: 1.04-1.18), with modest precision (Hartung-Knapp intervals crossed 1).
CONCLUSIONS: AI-enabled echocardiography provides a scalable digital biomarker that characterizes prominent epicardial adiposity and is associated with an adverse cardiometabolic profile, highlighting a role for echocardiography in metabolic risk stratification.
PMID:43066664 | DOI:10.1016/j.jacc.2026.10.066
European Heart Journal
© 2026 Imedsource, All Rights Reserved
For unrestricted access to content and a seamless experience,
Complete your profile to help us deliver a more personalized experience.