Back

Does explainable AI-ECG heart age differentiate pathological from physiological LV remodeling? A multi-cohort analysis including young elite athletes

Hempel, P.; Steinbrinker, T.; Graf, L.; Trivedi, S.; Singstad, B.-J.; Abela, M.; Niederseer, D.; Vollmer, M.; Dorr, M.; Spicher, N.; Krefting, D.

2025-09-08 sports medicine
10.1101/2025.09.06.25335008 medRxiv
Show abstract

AimArtificial intelligence applied to electrocardiography (AI-ECG) can derive a heart age or ECG-age, potentially reflecting waveform patterns that indicate cumulative myocardial stress. The heart age gap (HA-gap, {Delta}age) is defined as the difference between a persons ECG-age and chronological age. Former studies suggest a threshold of {Delta}age > 8 yrs as a biomarker for accelerated biological age, associated with higher risk for cardiovascular events. In this study, we investigate whether {Delta}age differentiates training-induced physiological from pathological left ventricular remodeling. MethodsAn AI-ECG was applied to 162 resting 12-lead ECGs of each professional footballers, population controls without cardiovascular disease, and patients with systolic heart failure (HF). Explainable AI identified contributing leads and waveforms, and results were compared with established ECG voltage criteria for left ventricular hypertrophy (Sokolow-Lyon, Cornell) and low QRS voltage (LQRSV). ResultsAccelerated HA ({Delta}age,+) was present in 38.9% of athletes, 35.8% of community controls, and 96.9% of HF patients. As a diagnostic criterion, accelerated HA achieved 96.9% sensitivity and 62.7% specificity for distinguishing diseased from healthy cohorts. In contrast, classical ECG voltage criteria showed lower sensitivity (6-17%) but higher specificity (85-100%). Correlation analyses confirmed significant associations of HA-gap with Cornell voltage ({rho} = 0.25, p < 0.001) and LQRSV (limb:{rho} = -0.43, p < 0.001; precordial:{rho} = -0.32, p < 0.001). ConclusionsThe AI-based HA-gap is a multi-factorial marker of ventricular remodeling beyond mass and can separate benign athletic hypertrophy from pathological remodeling with high sensitivity. Incorporating athlete and youth cohorts into model development could further improve specificity to enable future application in preventive and sports cardiology.

Matching journals

The top 1 journal accounts for 50% of the predicted probability mass.

1
Frontiers in Cardiovascular Medicine
49 papers in training set
Top 0.1%
54.4%
50% of probability mass above
2
Journal of the American Heart Association
119 papers in training set
Top 0.7%
9.5%
3
Scientific Reports
3102 papers in training set
Top 12%
7.1%
4
European Heart Journal - Digital Health
15 papers in training set
Top 0.1%
4.1%
5
PLOS ONE
4510 papers in training set
Top 47%
2.2%
6
European Journal of Epidemiology
40 papers in training set
Top 0.3%
1.7%
7
PLOS Computational Biology
1633 papers in training set
Top 18%
1.4%
8
Frontiers in Physiology
93 papers in training set
Top 3%
1.4%
9
Circulation
66 papers in training set
Top 2%
1.3%
10
Frontiers in Genetics
197 papers in training set
Top 7%
1.2%
11
Computers in Biology and Medicine
120 papers in training set
Top 3%
1.2%
12
Physiological Reports
35 papers in training set
Top 0.8%
1.0%
13
European Heart Journal
16 papers in training set
Top 0.6%
0.9%
14
Communications Biology
886 papers in training set
Top 17%
0.9%
15
Journal of Clinical Medicine
91 papers in training set
Top 6%
0.8%
16
Journal of Molecular and Cellular Cardiology
39 papers in training set
Top 0.9%
0.7%
17
Nature Cardiovascular Research
28 papers in training set
Top 0.6%
0.7%
18
JACC: Clinical Electrophysiology
11 papers in training set
Top 0.4%
0.7%
19
American Journal of Physiology-Heart and Circulatory Physiology
32 papers in training set
Top 1%
0.7%
20
BMC Cardiovascular Disorders
14 papers in training set
Top 2%
0.5%