1
|
Khurshid S, Churchill TW, Diamant N, Di Achille P, Reeder C, Singh P, Friedman SF, Wasfy MM, Alba GA, Maron BA, Systrom DM, Wertheim BM, Ellinor PT, Ho JE, Baggish AL, Batra P, Lubitz SA, Guseh JS. Deep learned representations of the resting 12-lead electrocardiogram to predict at peak exercise. Eur J Prev Cardiol 2024; 31:252-262. [PMID: 37798122 PMCID: PMC10809171 DOI: 10.1093/eurjpc/zwad321] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 07/19/2023] [Revised: 09/14/2023] [Accepted: 09/29/2023] [Indexed: 10/07/2023]
Abstract
AIMS To leverage deep learning on the resting 12-lead electrocardiogram (ECG) to estimate peak oxygen consumption (V˙O2peak) without cardiopulmonary exercise testing (CPET). METHODS AND RESULTS V ˙ O 2 peak estimation models were developed in 1891 individuals undergoing CPET at Massachusetts General Hospital (age 45 ± 19 years, 38% female) and validated in a separate test set (MGH Test, n = 448) and external sample (BWH Test, n = 1076). Three penalized linear models were compared: (i) age, sex, and body mass index ('Basic'), (ii) Basic plus standard ECG measurements ('Basic + ECG Parameters'), and (iii) basic plus 320 deep learning-derived ECG variables instead of ECG measurements ('Deep ECG-V˙O2'). Associations between estimated V˙O2peak and incident disease were assessed using proportional hazards models within 84 718 primary care patients without CPET. Inference ECGs preceded CPET by 7 days (median, interquartile range 27-0 days). Among models, Deep ECG-V˙O2 was most accurate in MGH Test [r = 0.845, 95% confidence interval (CI) 0.817-0.870; mean absolute error (MAE) 5.84, 95% CI 5.39-6.29] and BWH Test (r = 0.552, 95% CI 0.509-0.592, MAE 6.49, 95% CI 6.21-6.67). Deep ECG-V˙O2 also outperformed the Wasserman, Jones, and FRIEND reference equations (P < 0.01 for comparisons of correlation). Performance was higher in BWH Test when individuals with heart failure (HF) were excluded (r = 0.628, 95% CI 0.567-0.682; MAE 5.97, 95% CI 5.57-6.37). Deep ECG-V˙O2 estimated V˙O2peak <14 mL/kg/min was associated with increased risks of incident atrial fibrillation [hazard ratio 1.36 (95% CI 1.21-1.54)], myocardial infarction [1.21 (1.02-1.45)], HF [1.67 (1.49-1.88)], and death [1.84 (1.68-2.03)]. CONCLUSION Deep learning-enabled analysis of the resting 12-lead ECG can estimate exercise capacity (V˙O2peak) at scale to enable efficient cardiovascular risk stratification.
Collapse
Affiliation(s)
- Shaan Khurshid
- Cardiovascular Research Center, Massachusetts General Hospital, 185 Cambridge Street Suite 3201, Boston, MA 02114, USA
- Demoulas Center for Cardiac Arrhythmias, Division of Cardiology, Massachusetts General Hospital, 55 Fruit Street, GRB 109, Boston, MA 02114, USA
- Cardiovascular Disease Initiative, Broad Institute of Harvard and the Massachusetts Institute of Technology, 415 Main Street, Cambridge, MA 02142, USA
| | - Timothy W Churchill
- Cardiovascular Research Center, Massachusetts General Hospital, 185 Cambridge Street Suite 3201, Boston, MA 02114, USA
- Cardiovascular Performance Program, Division of Cardiology, Mass General Sports Medicine, Massachusetts General Hospital, 55 Fruit Street, GRB 109, Boston, MA 02114, USA
| | - Nathaniel Diamant
- Data Sciences Platform, Broad Institute of Harvard and the Massachusetts Institute of Technology, Cambridge, MA 02142, USA
| | - Paolo Di Achille
- Data Sciences Platform, Broad Institute of Harvard and the Massachusetts Institute of Technology, Cambridge, MA 02142, USA
| | - Christopher Reeder
- Data Sciences Platform, Broad Institute of Harvard and the Massachusetts Institute of Technology, Cambridge, MA 02142, USA
| | - Pulkit Singh
- Data Sciences Platform, Broad Institute of Harvard and the Massachusetts Institute of Technology, Cambridge, MA 02142, USA
| | - Samuel F Friedman
- Data Sciences Platform, Broad Institute of Harvard and the Massachusetts Institute of Technology, Cambridge, MA 02142, USA
| | - Meagan M Wasfy
- Cardiovascular Research Center, Massachusetts General Hospital, 185 Cambridge Street Suite 3201, Boston, MA 02114, USA
- Cardiovascular Performance Program, Division of Cardiology, Mass General Sports Medicine, Massachusetts General Hospital, 55 Fruit Street, GRB 109, Boston, MA 02114, USA
| | - George A Alba
- Division of Pulmonary and Critical Care Medicine, Massachusetts General Hospital, Boston, MA 02114, USA
| | - Bradley A Maron
- Division of Cardiovascular Medicine, Brigham and Women’s Hospital, Boston, MA 02115, USA
- Department of Medicine, University of Maryland School of Medicine, Baltimore, MD 21201, USA
- University of Maryland, Institute for Health Computing, Bethesda, MD, USA
| | - David M Systrom
- Division of Pulmonary and Critical Care Medicine, Brigham and Women’s Hospital, Boston, MA 02115, USA
| | - Bradley M Wertheim
- Division of Pulmonary and Critical Care Medicine, Brigham and Women’s Hospital, Boston, MA 02115, USA
| | - Patrick T Ellinor
- Cardiovascular Research Center, Massachusetts General Hospital, 185 Cambridge Street Suite 3201, Boston, MA 02114, USA
- Demoulas Center for Cardiac Arrhythmias, Division of Cardiology, Massachusetts General Hospital, 55 Fruit Street, GRB 109, Boston, MA 02114, USA
- Cardiovascular Disease Initiative, Broad Institute of Harvard and the Massachusetts Institute of Technology, 415 Main Street, Cambridge, MA 02142, USA
| | - Jennifer E Ho
- Division of Cardiology, Department of Medicine, Beth Israel Deaconess Medical Center, CardioVascular Institute, Boston, MA, USA
| | - Aaron L Baggish
- Cardiovascular Research Center, Massachusetts General Hospital, 185 Cambridge Street Suite 3201, Boston, MA 02114, USA
- Cardiovascular Performance Program, Division of Cardiology, Mass General Sports Medicine, Massachusetts General Hospital, 55 Fruit Street, GRB 109, Boston, MA 02114, USA
- Département Coeur-Vaisseaux, Le Centre Hospitalier Universitaire Vaudois (CHUV), Institut des Sciences du Sport, Université de Lausanne, Écublens, Vaud, Switzerland
| | - Puneet Batra
- Data Sciences Platform, Broad Institute of Harvard and the Massachusetts Institute of Technology, Cambridge, MA 02142, USA
| | - Steven A Lubitz
- Cardiovascular Research Center, Massachusetts General Hospital, 185 Cambridge Street Suite 3201, Boston, MA 02114, USA
- Demoulas Center for Cardiac Arrhythmias, Division of Cardiology, Massachusetts General Hospital, 55 Fruit Street, GRB 109, Boston, MA 02114, USA
- Cardiovascular Disease Initiative, Broad Institute of Harvard and the Massachusetts Institute of Technology, 415 Main Street, Cambridge, MA 02142, USA
| | - J Sawalla Guseh
- Cardiovascular Research Center, Massachusetts General Hospital, 185 Cambridge Street Suite 3201, Boston, MA 02114, USA
- Cardiovascular Performance Program, Division of Cardiology, Mass General Sports Medicine, Massachusetts General Hospital, 55 Fruit Street, GRB 109, Boston, MA 02114, USA
| |
Collapse
|