{"doi":"10.1093/eurjpc/zwad321","title":"Deep learned representations of the resting 12-lead electrocardiogram to predict at peak exercise","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.","journal":"European Journal of Preventive Cardiology","year":2023,"id":343371,"datarank":0.53705878956116,"base_score":2.70805020110221,"endowment":2.70805020110221,"self_citation_contribution":0.40620753016533157,"citation_network_contribution":0.13085125939582848,"self_endowment_contribution":0.40620753016533157,"citer_contribution":0.13085125939582848,"corpus_percentile":null,"corpus_rank":null,"citation_count":14,"citer_count":11,"citers_with_citation_signal":5,"citers_with_endowment":5,"datacite_reuse_total":0,"is_dataset":false,"is_dataset_confidence":0.7956,"is_data_producer":false,"deposit_databanks":null,"is_oa":true,"file_count":0,"downloads":0,"has_version_chain":false,"published_date":"2023-01-01","fair_score":null,"fair_percentile":null,"algorithm_id":"datarank_citation_only_1hop_v6","ranking_scope":"data_only","authors":[{"id":316988,"name":"Timothy W. Churchill","orcid":"0000-0002-0215-3049","position":1,"is_corresponding":false},{"id":18141,"name":"Nathaniel Diamant","orcid":"0000-0002-1738-304X","position":2,"is_corresponding":false},{"id":464102,"name":"Paolo Di Achille","orcid":"0000-0001-9256-0678","position":3,"is_corresponding":false},{"id":810201,"name":"Christopher Reeder","orcid":"0000-0002-3893-2423","position":4,"is_corresponding":false},{"id":810202,"name":"Pulkit Singh","orcid":"0000-0002-1538-5519","position":5,"is_corresponding":false},{"id":236909,"name":"Sam Friedman","orcid":"0000-0002-0688-2169","position":6,"is_corresponding":false},{"id":432286,"name":"Meagan M. Wasfy","orcid":"0000-0003-0398-0481","position":7,"is_corresponding":false},{"id":236776,"name":"George A. Alba","orcid":"0000-0002-6876-7836","position":8,"is_corresponding":false},{"id":87929,"name":"Bradley A. Maron","orcid":"0000-0002-6784-764X","position":9,"is_corresponding":false},{"id":377392,"name":"David M. Systrom","orcid":"0000-0002-9610-6330","position":10,"is_corresponding":false},{"id":236775,"name":"Bradley M. Wertheim","orcid":"0000-0002-1414-4692","position":11,"is_corresponding":false},{"id":896,"name":"Patrick T. Ellinor","orcid":"0000-0002-2067-0533","position":12,"is_corresponding":false},{"id":892,"name":"Jennifer E. Ho","orcid":"0000-0002-7987-4768","position":13,"is_corresponding":false},{"id":255090,"name":"Aaron L. Baggish","orcid":"0000-0003-2042-1489","position":14,"is_corresponding":false},{"id":552484,"name":"Puneet Batra","orcid":"0000-0001-6822-0593","position":15,"is_corresponding":false},{"id":1083,"name":"Steven A. Lubitz","orcid":"0000-0002-9599-4866","position":16,"is_corresponding":false},{"id":481021,"name":"J. Sawalla Guseh","orcid":"0000-0003-0992-5635","position":17,"is_corresponding":false},{"id":552483,"name":"Shaan Khurshid","orcid":"0000-0002-2840-4539","position":0,"is_corresponding":true}],"reference_count":54,"raw_metadata":{"citation_network_status":"fetched"},"created_at":"2026-07-19T01:11:21.758449Z","pmid":"37798122","pmcid":null,"fwci":null,"citation_percentile":null,"influential_citations":0,"oa_status":null,"license":null,"views":0,"total_file_size_bytes":0,"version_count":0,"fair_f":null,"fair_a":null,"fair_i":null,"fair_r":null,"fair_zscore":null,"fair_rationale":null,"fair_model":null,"fair_agent_version":null,"fair_fulltext_source":null,"fair_has_llm":null,"fair_computed_at":null,"clinical_trials":[],"software_tools":[],"db_accessions":[],"linked_datasets":[],"topics":[]}