{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"institution":[{"name":"medRxiv"}],"indexed":{"date-parts":[[2026,9,21]],"date-time":"2026-09-21T16:14:12Z","timestamp":1790007252965,"version":"4.0.1"},"posted":{"date-parts":[[2025,7,2]]},"group-title":"Cardiovascular Medicine","reference-count":55,"publisher":"openRxiv","license":[{"start":{"date-parts":[[2025,7,2]],"date-time":"2025-07-02T00:00:00Z","timestamp":1751414400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.medrxiv.org\/about\/FAQ#license"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"accepted":{"date-parts":[[2025,8,12]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                <jats:sec id=\"s1\">\n                  <jats:title>Introduction<\/jats:title>\n                  <jats:p>Artificial intelligence (AI)-enhanced electrocardiogram (ECG) models are designed to detect specific anatomical and functional cardiac abnormalities. Understanding the selectivity of their phenotypic associations is essential to inform their clinical use. Here, we sought to assess whether AI-ECG models function as condition-specific classifiers or broader cardiovascular risk markers.<\/jats:p>\n                <\/jats:sec>\n                <jats:sec id=\"s2\">\n                  <jats:title>Methods<\/jats:title>\n                  <jats:p>We included four distinct study populations, drawn from both electronic health records (EHR) and prospective cohort studies. We deployed six image-based AI-ECG models, including five validated models for the detection of left ventricular systolic dysfunction (LVSD), aortic stenosis (AS), mitral regurgitation (MR), left ventricular hypertrophy (LVH), a composite model for structural heart disease (SHD), and a negative control AI-ECG model for biological sex. Additionally, we developed six experimental models designed to identify non-cardiovascular conditions. Diagnosis codes from EHR and cohorts were transformed into interpretable phenotypes using a phenome-wide association study (PheWAS) framework. We assessed associations of AI-ECG probabilities with cross-sectional phenotypes using logistic regression, and with new-onset cardiovascular diseases using Cox regression. Pearson correlation coefficients were calculated to compare phenotypic signatures.<\/jats:p>\n                <\/jats:sec>\n                <jats:sec id=\"s3\">\n                  <jats:title>Results<\/jats:title>\n                  <jats:p>The study included one random ECG from 233,689 individuals (mean age 59\u00b118 years, 130,084 [56%] women) across sites. Each of the five AI-ECG models was more likely to be associated with cardiovascular phenotypes compared with other phenotype groups (odds ratios ranging from 2.16 to 4.41, p&lt;10\u25a1), while the sex model did not show a similar pattern. All AI-ECG models were significantly associated with their respective target phenotype, but also showed similar or stronger associations with a broad range of other cardiovascular phenotypes. Phenotypic associations were similar across AI-ECG models trained for different conditions, which was not observed in models for non-cardiovascular conditions. Correlation of phenotype association patterns between models was high (r = 0.65\u20130.99). This pattern was consistent across all models, external datasets, and in both cross-sectional and prospective analyses.<\/jats:p>\n                <\/jats:sec>\n                <jats:sec id=\"s4\">\n                  <jats:title>Conclusions<\/jats:title>\n                  <jats:p>Despite being developed to detect specific cardiovascular conditions, AI-ECG models detect the presence and predict the future development of a broad range of cardiovascular diseases with similar propensity. This challenges their role as binary diagnostic tools and instead supports their use as broader cardiovascular biomarkers.<\/jats:p>\n                <\/jats:sec>","DOI":"10.1101\/2025.07.01.25330665","type":"posted-content","created":{"date-parts":[[2025,7,2]],"date-time":"2025-07-02T13:20:19Z","timestamp":1751462419000},"source":"Crossref","is-referenced-by-count":1,"title":["Phenotypic Selectivity of Artificial Intelligence-enhanced Electrocardiography in Cardiovascular Diagnosis and Risk Prediction"],"prefix":"10.64898","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-8082-0161","authenticated-orcid":false,"given":"Philip M.","family":"Croon","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5664-4126","authenticated-orcid":false,"given":"Lovedeep S.","family":"Dhingra","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9141-5188","authenticated-orcid":false,"given":"Dhruva","family":"Biswas","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4362-0720","authenticated-orcid":false,"given":"Evangelos K","family":"Oikonomou","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9467-6199","authenticated-orcid":false,"given":"Rohan","family":"Khera","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"54368","reference":[{"key":"2025081404100668000_2025.07.01.25330665v2.1","doi-asserted-by":"crossref","first-page":"465","DOI":"10.1038\/s41569-020-00503-2","article-title":"Artificial intelligence-enhanced electrocardiography in cardiovascular disease management","volume":"18","year":"2021","journal-title":"Nat Rev Cardiol"},{"key":"2025081404100668000_2025.07.01.25330665v2.2","doi-asserted-by":"crossref","first-page":"1179","DOI":"10.1093\/europace\/euaa377","article-title":"Deep learning and the electrocardiogram: review of the current state-of-the-art","volume":"23","year":"2021","journal-title":"Europace"},{"key":"2025081404100668000_2025.07.01.25330665v2.3","doi-asserted-by":"publisher","DOI":"10.1161\/CIRCULATIONAHA.122.062646"},{"key":"2025081404100668000_2025.07.01.25330665v2.4","first-page":"20552076231187250","article-title":"Mortality risk prediction of the electrocardiogram as an informative indicator of cardiovascular diseases","volume":"9","year":"2023","journal-title":"Digit Health"},{"key":"2025081404100668000_2025.07.01.25330665v2.5","doi-asserted-by":"publisher","DOI":"10.1038\/s41591-018-0240-2"},{"key":"2025081404100668000_2025.07.01.25330665v2.6","doi-asserted-by":"crossref","first-page":"151","DOI":"10.1016\/j.jelectrocard.2020.02.008","article-title":"Artificial intelligence for detecting mitral regurgitation using electrocardiography","volume":"59","year":"2020","journal-title":"J Electrocardiol"},{"key":"2025081404100668000_2025.07.01.25330665v2.7","doi-asserted-by":"crossref","first-page":"412","DOI":"10.1093\/europace\/euz324","article-title":"Comparing the performance of artificial intelligence and conventional diagnosis criteria for detecting left ventricular hypertrophy using electrocardiography","volume":"22","year":"2020","journal-title":"Europace"},{"key":"2025081404100668000_2025.07.01.25330665v2.8","doi-asserted-by":"crossref","first-page":"1044","DOI":"10.1093\/eurheartj\/ehae914","article-title":"Heart failure risk stratification using artificial intelligence applied to electrocardiogram images: a multinational study","volume":"46","year":"2025","journal-title":"Eur Heart J"},{"key":"2025081404100668000_2025.07.01.25330665v2.9","doi-asserted-by":"publisher","DOI":"10.1016\/s0140-6736(22)01637-3"},{"key":"2025081404100668000_2025.07.01.25330665v2.10","doi-asserted-by":"publisher","DOI":"10.1016\/S2589-7500(24)00172-9"},{"key":"2025081404100668000_2025.07.01.25330665v2.11","doi-asserted-by":"crossref","first-page":"214","DOI":"10.1001\/jamacardio.2024.4796","article-title":"Artificial intelligence-enhanced electrocardiography for prediction of incident hypertension","volume":"10","year":"2025","journal-title":"JAMA Cardiol"},{"key":"2025081404100668000_2025.07.01.25330665v2.12","unstructured":"Anumana ECG-AI LEF [Internet]. [cited 2025 Feb 11];Available from: https:\/\/anumana.ai\/ecg-ai-lef"},{"key":"2025081404100668000_2025.07.01.25330665v2.13","doi-asserted-by":"publisher","DOI":"10.1016\/j.amjcard.2012.09.029"},{"key":"2025081404100668000_2025.07.01.25330665v2.14","doi-asserted-by":"crossref","first-page":"11289","DOI":"10.1038\/s41598-024-62142-8","article-title":"Association of major and minor ECG abnormalities with traditional cardiovascular risk factors in the general population: a large scale study","volume":"14","year":"2024","journal-title":"Sci Rep"},{"key":"2025081404100668000_2025.07.01.25330665v2.15","doi-asserted-by":"publisher","DOI":"10.1016\/j.jacc.2014.05.050"},{"key":"2025081404100668000_2025.07.01.25330665v2.16","doi-asserted-by":"crossref","first-page":"1443","DOI":"10.1016\/j.amjcard.2019.01.050","article-title":"Relation of minor electrocardiographic abnormalities to cardiovascular mortality","volume":"123","year":"2019","journal-title":"Am J Cardiol"},{"key":"2025081404100668000_2025.07.01.25330665v2.17","doi-asserted-by":"crossref","unstructured":"Croon PM , Pedroso AF , Khera R . The emerging role of AI in transforming cardiovascular care. Future Cardiol. 2025;1\u20134.","DOI":"10.1080\/14796678.2025.2492973"},{"key":"2025081404100668000_2025.07.01.25330665v2.18","doi-asserted-by":"publisher","DOI":"10.1016\/j.echo.2017.01.007"},{"key":"2025081404100668000_2025.07.01.25330665v2.19","doi-asserted-by":"crossref","first-page":"1302","DOI":"10.1016\/j.jacc.2025.01.030","article-title":"Ensemble deep learning algorithm for structural heart disease screening using electrocardiographic images","volume":"85","year":"2025","journal-title":"J Am Coll Cardiol"},{"key":"2025081404100668000_2025.07.01.25330665v2.20","doi-asserted-by":"crossref","first-page":"855","DOI":"10.1093\/jamia\/ocae002","article-title":"Biometric contrastive learning for data-efficient deep learning from electrocardiographic images","volume":"31","year":"2024","journal-title":"J Am Med Inform Assoc"},{"key":"2025081404100668000_2025.07.01.25330665v2.21","doi-asserted-by":"crossref","first-page":"317","DOI":"10.1177\/8756479319836781","article-title":"Diastolic dysfunction and the new recommendations for echocardiographic assessment of left ventricular diastolic function: Summary of guidelines and novelties in diagnosis and grading","volume":"35","year":"2019","journal-title":"J Diagn Med Sonogr"},{"key":"2025081404100668000_2025.07.01.25330665v2.22","doi-asserted-by":"publisher","DOI":"10.1161\/CIRCEP.119.007284"},{"key":"2025081404100668000_2025.07.01.25330665v2.23","doi-asserted-by":"publisher","DOI":"10.1093\/bioinformatics\/btq126"},{"key":"2025081404100668000_2025.07.01.25330665v2.24","doi-asserted-by":"publisher","DOI":"10.1186\/s13040-019-0203-2"},{"key":"2025081404100668000_2025.07.01.25330665v2.25","doi-asserted-by":"crossref","first-page":"5711316","DOI":"10.1155\/cdr\/5711316","article-title":"Exploring potential drug targets in multiple cardiovascular diseases: A study based on proteome-wide Mendelian randomization and colocalization analysis","volume":"2025","year":"2025","journal-title":"Cardiovasc Ther"},{"key":"2025081404100668000_2025.07.01.25330665v2.26","first-page":"ooaf006","article-title":"Exploring beyond diagnoses in electronic health records to improve discovery: a review of the phenome-wide association study","volume":"8","year":"2025","journal-title":"JAMIA Open"},{"key":"2025081404100668000_2025.07.01.25330665v2.27","doi-asserted-by":"publisher","DOI":"10.1038\/nbt.2749"},{"key":"2025081404100668000_2025.07.01.25330665v2.28","unstructured":"PheWAS Resources [Internet]. [cited 2025 Mar 21];Available from: https:\/\/phewascatalog.org\/phewas\/#phe12"},{"key":"2025081404100668000_2025.07.01.25330665v2.29","doi-asserted-by":"publisher","DOI":"10.1016\/j.ajhg.2010.03.003"},{"key":"2025081404100668000_2025.07.01.25330665v2.30","doi-asserted-by":"crossref","first-page":"e14325","DOI":"10.2196\/14325","article-title":"Mapping ICD-10 and ICD-10-CM codes to phecodes: Workflow development and initial evaluation","volume":"7","year":"2019","journal-title":"JMIR Med Inform"},{"key":"2025081404100668000_2025.07.01.25330665v2.31","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0175508"},{"key":"2025081404100668000_2025.07.01.25330665v2.32","doi-asserted-by":"crossref","first-page":"626","DOI":"10.1093\/ehjdh\/ztab080","article-title":"ECG-AI: electrocardiographic artificial intelligence model for prediction of heart failure","volume":"2","year":"2021","journal-title":"Eur Heart J Digit Health"},{"key":"2025081404100668000_2025.07.01.25330665v2.33","doi-asserted-by":"crossref","unstructured":"Sangha V , Dhingra LS , Aminorroaya A , Croon PM , Sikand NV , Sen S , Martinez MW , Maron MS , Krumholz HM , Asselbergs FW , Oikonomou EK , Khera R . Identification of hypertrophic cardiomyopathy on electrocardiographic images with deep learning. Nat Cardiovasc Res. 2025;1\u201310.","DOI":"10.1038\/s44161-025-00685-3"},{"key":"2025081404100668000_2025.07.01.25330665v2.34","doi-asserted-by":"publisher","DOI":"10.1093\/bioinformatics\/btu197"},{"key":"2025081404100668000_2025.07.01.25330665v2.35","unstructured":"PheWAS: The PheWAS R package [Internet]. Github; [cited 2025 Feb 12]. Available from: https:\/\/github.com\/PheWAS\/PheWAS"},{"key":"2025081404100668000_2025.07.01.25330665v2.36","doi-asserted-by":"publisher","DOI":"10.1038\/s41591-021-01335-4"},{"key":"2025081404100668000_2025.07.01.25330665v2.37","doi-asserted-by":"crossref","unstructured":"Oikonomou EK , Sangha V , Shankar SV , Coppi A , Krumholz HM , Nasir K , Miller EJ , Gallegos-Kattan C , Al-Kindi S , Khera R . Tracking the pre-clinical progression of transthyretin amyloid cardiomyopathy using artificial intelligence-enabled electrocardiography and echocardiography. medRxiv [Internet]. 2024 [cited 2024 Oct 9];Available from: https:\/\/pubmed.ncbi.nlm.nih.gov\/39252891\/","DOI":"10.1101\/2024.08.25.24312556"},{"key":"2025081404100668000_2025.07.01.25330665v2.38","doi-asserted-by":"crossref","first-page":"2733","DOI":"10.1007\/s11255-022-03165-w","article-title":"Artificial intelligence assessment for early detection and prediction of renal impairment using electrocardiography","volume":"54","year":"2022","journal-title":"Int Urol Nephrol"},{"key":"2025081404100668000_2025.07.01.25330665v2.39","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejim.2025.04.036"},{"key":"2025081404100668000_2025.07.01.25330665v2.40","doi-asserted-by":"crossref","first-page":"322","DOI":"10.1038\/s41746-025-01727-7","article-title":"Determinants of artificial intelligence electrocardiogram-derived age and its association with cardiovascular events and mortality: a systematic review and meta-analysis","volume":"8","year":"2025","journal-title":"NPJ Digit Med"},{"key":"2025081404100668000_2025.07.01.25330665v2.41","doi-asserted-by":"publisher","DOI":"10.1056\/aioa2400421"},{"key":"2025081404100668000_2025.07.01.25330665v2.42","doi-asserted-by":"crossref","first-page":"1314","DOI":"10.1016\/j.jacc.2025.02.008","article-title":"Implementing ECG-AI in clinical care: From pixel to practice","volume":"85","year":"2025","journal-title":"J Am Coll Cardiol"},{"key":"2025081404100668000_2025.07.01.25330665v2.43","doi-asserted-by":"crossref","unstructured":"Friedman SF , Khurshid S , Venn RA , Wang X , Diamant N , Di Achille P , Weng L-C , Choi SH , Reeder C , Pirruccello JP , Singh P , Lau ES , Philippakis A , Anderson CD , Maddah M , Batra P , Ellinor PT , Ho JE , Lubitz SA . Unsupervised deep learning of electrocardiograms enables scalable human disease profiling. NPJ Digit Med. 2025;8:23.","DOI":"10.1038\/s41746-024-01418-9"},{"key":"2025081404100668000_2025.07.01.25330665v2.44","doi-asserted-by":"crossref","unstructured":"Hughes JW , Theurer J , Vukadinovic M , Rogers AJ , Somani S , Kang G , Ghazizadeh Z , O\u2019Sullivan JW , Jain SS , Gomes B , Salerno M , Ashley E , Zou JY , Perez MV , Ouyang D . A deep learning phenome wide association study of the electrocardiogram. Eur Heart J Digit Health. 2025;ztaf047.","DOI":"10.1093\/ehjdh\/ztaf047"},{"key":"2025081404100668000_2025.07.01.25330665v2.45","doi-asserted-by":"crossref","first-page":"30","DOI":"10.1016\/j.jelectrocard.2024.01.006","article-title":"Artificial intelligence-enhanced electrocardiography for accurate diagnosis and management of cardiovascular diseases","volume":"83","year":"2024","journal-title":"J Electrocardiol"},{"key":"2025081404100668000_2025.07.01.25330665v2.46","doi-asserted-by":"publisher","DOI":"10.1161\/hypertensionaha.119.14240"},{"key":"2025081404100668000_2025.07.01.25330665v2.47","doi-asserted-by":"publisher","DOI":"10.1007\/s40620-020-00842-w"},{"key":"2025081404100668000_2025.07.01.25330665v2.48","doi-asserted-by":"crossref","first-page":"1054","DOI":"10.1093\/eurheartj\/ehae906","article-title":"Using artificial intelligence to spot heart failure from ECGs: is it prime time?","volume":"46","year":"2025","journal-title":"Eur Heart J"},{"key":"2025081404100668000_2025.07.01.25330665v2.49","doi-asserted-by":"publisher","DOI":"10.1093\/eurheartj\/ehaf387"},{"key":"2025081404100668000_2025.07.01.25330665v2.50","doi-asserted-by":"crossref","first-page":"100998","DOI":"10.1016\/j.jacadv.2024.100998","article-title":"Impact of case and control selection on training artificial intelligence screening of cardiac amyloidosis","volume":"3","year":"2024","journal-title":"JACC Adv"},{"key":"2025081404100668000_2025.07.01.25330665v2.51","doi-asserted-by":"crossref","first-page":"328","DOI":"10.1038\/s41746-025-01717-9","article-title":"A general framework for governing marketed AI\/ML medical devices","volume":"8","year":"2025","journal-title":"NPJ Digit Med"},{"key":"2025081404100668000_2025.07.01.25330665v2.52","doi-asserted-by":"crossref","first-page":"e258052","DOI":"10.1001\/jamanetworkopen.2025.8052","article-title":"Generalizability of FDA-approved AI-enabled medical devices for clinical use","volume":"8","year":"2025","journal-title":"JAMA Netw Open"},{"key":"2025081404100668000_2025.07.01.25330665v2.53","doi-asserted-by":"publisher","DOI":"10.1001\/jama.2021.20356"},{"key":"2025081404100668000_2025.07.01.25330665v2.54","doi-asserted-by":"publisher","DOI":"10.1101\/2025.07.08.25331129"},{"key":"2025081404100668000_2025.07.01.25330665v2.55","doi-asserted-by":"crossref","unstructured":"Sau A , Zeidaabadi B , Patlatzoglou K , Pastika L , Ribeiro AH , Sabino E , Peters NS , Ribeiro ALP , Kramer DB , Waks JW , Ng FS . A comparison of artificial intelligence-enhanced electrocardiography approaches for prediction of time-to-mortality using electrocardiogram images. Eur Heart J Digit Health. 2024;ztae090.","DOI":"10.1093\/ehjdh\/ztae090"}],"container-title":[],"original-title":[],"link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.1101\/2025.07.01.25330665","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,1,15]],"date-time":"2026-01-15T20:39:43Z","timestamp":1768509583000},"score":1,"resource":{"primary":{"URL":"http:\/\/medrxiv.org\/lookup\/doi\/10.1101\/2025.07.01.25330665"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,7,2]]},"references-count":55,"URL":"https:\/\/doi.org\/10.1101\/2025.07.01.25330665","relation":{"is-preprint-of":[{"id-type":"doi","id":"10.1161\/CIRCULATIONAHA.125.076279","asserted-by":"subject"}]},"subject":[],"published":{"date-parts":[[2025,7,2]]},"subtype":"preprint"}}