{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"institution":[{"name":"medRxiv"}],"indexed":{"date-parts":[[2026,8,14]],"date-time":"2026-08-14T21:29:36Z","timestamp":1786742976785,"version":"build-2736575974"},"posted":{"date-parts":[[2026,8,12]]},"group-title":"Health Informatics","reference-count":27,"publisher":"openRxiv","license":[{"start":{"date-parts":[[2026,8,12]],"date-time":"2026-08-12T00:00:00Z","timestamp":1786492800000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"funder":[{"name":"National Center for Advancing Translational Sciences of the National Institutes of Health","award":["UL1TR001878"],"award-info":[{"award-number":["UL1TR001878"]}]},{"id":[{"id":"https:\/\/ror.org\/041phzg73","id-type":"ROR","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"accepted":{"date-parts":[[2026,8,12]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                <jats:p>Hypertrophic and dilated cardiomyopathy (HCM and DCM) carry substantial morbidity and mortality, yet diagnosis may be delayed, particularly when presentation is nonspecific. Existing machine-learning approaches to cardiomyopathy phenotyping, genotype prediction, and risk stratification commonly rely on disease-specific, hand-engineered features drawn from echocardiography, cardiac MRI, ECG, or curated clinical variables. We evaluated whether a general-purpose clinical foundation model, CLMBR-T-base, pre-trained via next-clinical-event prediction with no cardiomyopathy-specific supervision, could produce linearly separable embeddings for all three case\/control cohorts. Using EHR data from the Penn Medicine BioBank, we constructed cohorts for (1) prediction of a first recorded qualifying HCM\/DCM diagnosis at 1-, 3-, and 6-month horizons, decomposed into eventual-versus-never-case and imminent-versus-eventual comparisons; (2) genetic carrier status prediction among diagnosed patients with completed gene panels; and (3) prediction of heart-failure hospitalization, and all-cause mortality as both binary and time-to-event outcomes. Linear probes fitted to frozen embeddings achieved AUROCs of 0.75-0.82 for onset prediction, 0.74-0.75 for genotype status, and Harrell\u2019s concordance of 0.65-0.80 for time-to-event outcomes. Decomposing the onset prediction task reveals that the model often misclassifies patients who were diagnosed later as positive, suggesting the patient journey embeddings encode disease state more reliably than care timing. These results suggest that a single, generically pretrained EHR embedding can support multiple clinically motivated prediction problems in CM without disease-specific feature engineering.<\/jats:p>","DOI":"10.64898\/2026.08.10.26360107","type":"posted-content","created":{"date-parts":[[2026,8,12]],"date-time":"2026-08-12T22:40:14Z","timestamp":1786574414000},"source":"Crossref","is-referenced-by-count":0,"title":["Longitudinal Clinical Foundation Models Augmented with Genomics for Early Detection and Risk Stratification of Inherited Cardiomyopathy"],"prefix":"10.64898","author":[{"ORCID":"https:\/\/orcid.org\/0009-0005-7080-4877","authenticated-orcid":false,"given":"Andrew","family":"Zolensky","sequence":"first","affiliation":[{"name":"Division of Translational Medicine and Human Genetics, Department of Medicine, University of Pennsylvania, Pennsylvania, PA, 19104, USA"},{"name":"Division of Informatics, Department of Biostatistics, Epidemiology and Informatics, University of Pennsylvania, Pennsylvania, PA, 19104, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-3117-685X","authenticated-orcid":false,"given":"Colleen M.","family":"Kripke","sequence":"additional","affiliation":[{"name":"Division of Translational Medicine and Human Genetics, Department of Medicine, University of Pennsylvania, Pennsylvania, PA, 19104, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0945-5816","authenticated-orcid":false,"given":"Karl","family":"Keat","sequence":"additional","affiliation":[{"name":"Division of Translational Medicine and Human Genetics, Department of Medicine, University of Pennsylvania, Pennsylvania, PA, 19104, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8009-1632","authenticated-orcid":false,"given":"Scott M.","family":"Damrauer","sequence":"additional","affiliation":[{"name":"Department of Surgery, University of Pennsylvania, Pennsylvania, PA, 19104, USA"},{"name":"Department of Genetics, University of Pennsylvania, Pennsylvania, PA, 19104, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9937-9932","authenticated-orcid":false,"given":"Michael G.","family":"Levin","sequence":"additional","affiliation":[{"name":"Division of Cardiology, Department of Medicine, University of Pennsylvania, Pennsylvania, PA, 19104, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5063-9107","authenticated-orcid":false,"given":"Anurag","family":"Verma","sequence":"additional","affiliation":[{"name":"Division of Translational Medicine and Human Genetics, Department of Medicine, University of Pennsylvania, Pennsylvania, PA, 19104, USA"},{"name":"Division of Informatics, Department of Biostatistics, Epidemiology and Informatics, University of Pennsylvania, Pennsylvania, PA, 19104, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"54368","reference":[{"key":"2026081413401048000_2026.08.10.26360107v1.1","doi-asserted-by":"publisher","DOI":"10.3390\/ijms22147722"},{"key":"2026081413401048000_2026.08.10.26360107v1.2","doi-asserted-by":"publisher","DOI":"10.1161\/01.CIR.93.5.841"},{"key":"2026081413401048000_2026.08.10.26360107v1.3","doi-asserted-by":"crossref","first-page":"2467","DOI":"10.2174\/1389201011208062467","article-title":"Genetics of hypertrophic and dilated cardiomyopathy","volume":"13","year":"2012","journal-title":"Curr Pharm Biotechnol"},{"key":"2026081413401048000_2026.08.10.26360107v1.4","first-page":"758","article-title":"Rediscovering the value of exercise in patients with hypertrophic cardiomyopathy","volume":"51","year":"2022","journal-title":"Zhejiang Da Xue Xue Bao Yi Xue Ban"},{"key":"2026081413401048000_2026.08.10.26360107v1.5","doi-asserted-by":"publisher","DOI":"10.1093\/icvts\/ivs352"},{"key":"2026081413401048000_2026.08.10.26360107v1.6","doi-asserted-by":"crossref","first-page":"32","DOI":"10.1038\/s41572-019-0084-1","article-title":"Dilated cardiomyopathy","volume":"5","year":"2019","journal-title":"Nat Rev Dis Primers"},{"key":"2026081413401048000_2026.08.10.26360107v1.7","doi-asserted-by":"publisher","DOI":"10.1007\/s11897-024-00654-0"},{"key":"2026081413401048000_2026.08.10.26360107v1.8","doi-asserted-by":"publisher","DOI":"10.1080\/13696998.2023.2208966"},{"key":"2026081413401048000_2026.08.10.26360107v1.9","doi-asserted-by":"crossref","unstructured":"M. Wornow , R. Thapa , E. Steinberg , J. A. Fries and N. H. Shah , EHRSHOT: An EHR benchmark for few-shot evaluation of foundation models, arXiv preprint (2023), https:\/\/arxiv.org\/abs\/2307.02028.","DOI":"10.52202\/075280-2933"},{"key":"2026081413401048000_2026.08.10.26360107v1.10","doi-asserted-by":"publisher","DOI":"10.1016\/j.jbi.2020.103637"},{"key":"2026081413401048000_2026.08.10.26360107v1.11","unstructured":"A. Tomihari and I. Sato , Understanding linear probing then fine-tuning language models from NTK perspective Advances in Neural Information Processing Systems 37 (NeurIPS 2024) 2024. https:\/\/proceedings.neurips.cc\/paper_files\/paper\/2024\/file\/fcc22e5b7d5d2155d994da22d045f0a6-Paper-Conference.pdf."},{"key":"2026081413401048000_2026.08.10.26360107v1.12","unstructured":"B. Arnrich , E. Choi , J. A. Fries , M. B. A. McDermott , J. Oh , T. J. Pollard , N. Shah , E. Steinberg , M. Wornow and R. van de Water , Medical event data standard (MEDS): Facilitating machine learning for health, Workshop on Learning from Time Series for Health, ICLR 2024 (2024), https:\/\/openreview.net\/pdf?id=IsHy2ebjIG."},{"key":"2026081413401048000_2026.08.10.26360107v1.13","doi-asserted-by":"crossref","unstructured":"N. Z. Farahani , S. P. Arunachalam , D. S. B. Sundaram , K. Pasupathy , M. Enayati and A. M. Arruda-Olson , Explanatory analysis of a machine learning model to identify hypertrophic cardiomyopathy patients from EHR using diagnostic codes Proc IEEE Int Conf Bioinform Biomed 2020.","DOI":"10.1109\/BIBM49941.2020.9313231"},{"key":"2026081413401048000_2026.08.10.26360107v1.14","doi-asserted-by":"crossref","first-page":"102914","DOI":"10.1016\/j.jacadv.2026.102914","article-title":"Artificial intelligence for the detection of hypertrophic cardiomyopathy from standard electrocardiogram","volume":"5","year":"2026","journal-title":"JACC Adv"},{"key":"2026081413401048000_2026.08.10.26360107v1.15","doi-asserted-by":"crossref","first-page":"e003259","DOI":"10.1161\/CIRCGEN.120.003259","article-title":"Prediction of genotype positivity in patients with hypertrophic cardiomyopathy using machine learning","volume":"14","year":"2021","journal-title":"Circ Genom Precis Med"},{"key":"2026081413401048000_2026.08.10.26360107v1.16","doi-asserted-by":"publisher","DOI":"10.1161\/CIRCIMAGING.114.002411"},{"key":"2026081413401048000_2026.08.10.26360107v1.17","doi-asserted-by":"crossref","first-page":"3931","DOI":"10.1007\/s00330-020-07454-9","article-title":"Deep learning algorithm to improve hypertrophic cardiomyopathy mutation prediction using cardiac cine images","volume":"31","year":"2021","journal-title":"Eur Radiol"},{"key":"2026081413401048000_2026.08.10.26360107v1.18","doi-asserted-by":"crossref","first-page":"669860","DOI":"10.3389\/fcvm.2021.669860","article-title":"Deep learning analysis of echocardiographic images to predict positive genotype in patients with hypertrophic cardiomyopathy","volume":"8","year":"2021","journal-title":"Front Cardiovasc Med"},{"key":"2026081413401048000_2026.08.10.26360107v1.19","doi-asserted-by":"publisher","DOI":"10.1016\/j.mayocp.2014.01.025"},{"key":"2026081413401048000_2026.08.10.26360107v1.20","doi-asserted-by":"crossref","first-page":"1115","DOI":"10.1016\/j.jacc.2022.06.040","article-title":"Clinical risk score to predict pathogenic genotypes in patients with dilated cardiomyopathy","volume":"80","year":"2022","journal-title":"J Am Coll Cardiol"},{"key":"2026081413401048000_2026.08.10.26360107v1.21","doi-asserted-by":"crossref","unstructured":"T. Geyer , C. McIntosh , V. Sood , J. Matos , H. McKee and K. Hanneman , Machine learning\u2013based model for major adverse cardiac event prediction in patients with hypertrophic cardiomyopathy, Radiology: Cardiothoracic Imaging 8 (06 2026).","DOI":"10.1148\/ryct.250433"},{"key":"2026081413401048000_2026.08.10.26360107v1.22","first-page":"880","article-title":"Machine learning in hypertrophic cardiomyopathy: Nonlinear model from clinical and CMR features predicting cardiovascular events","volume":"17","year":"2024","journal-title":"JACC Cardiovasc Imaging"},{"key":"2026081413401048000_2026.08.10.26360107v1.23","doi-asserted-by":"publisher","DOI":"10.1038\/s44161-025-00679-1"},{"key":"2026081413401048000_2026.08.10.26360107v1.24","doi-asserted-by":"crossref","first-page":"20210259","DOI":"10.1259\/bjr.20210259","article-title":"A machine-learning-based method to predict adverse events in patients with dilated cardiomyopathy and severely reduced ejection fractions","volume":"94","year":"2021","journal-title":"Br J Radiol"},{"key":"2026081413401048000_2026.08.10.26360107v1.25","doi-asserted-by":"crossref","first-page":"922","DOI":"10.1007\/s10439-020-02639-1","article-title":"Machine learning outcome prediction in dilated cardiomyopathy using regional left ventricular multiparametric strain","volume":"49","year":"2021","journal-title":"Ann Biomed Eng"},{"key":"2026081413401048000_2026.08.10.26360107v1.26","doi-asserted-by":"crossref","first-page":"e0297793","DOI":"10.1371\/journal.pone.0297793","article-title":"Deep learning-based prediction of major arrhythmic events in dilated cardiomyopathy: A proof of concept study","volume":"19","year":"2024","journal-title":"PLoS One"},{"key":"2026081413401048000_2026.08.10.26360107v1.27","doi-asserted-by":"crossref","first-page":"1974","DOI":"10.3390\/jpm12121974","article-title":"The Penn Medicine BioBank: Towards a genomics-enabled learning healthcare system to accelerate precision medicine in a diverse population","volume":"12","year":"2022","journal-title":"J Pers Med"}],"container-title":[],"original-title":[],"link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.64898\/2026.08.10.26360107","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,14]],"date-time":"2026-08-14T20:40:18Z","timestamp":1786740018000},"score":1,"resource":{"primary":{"URL":"http:\/\/medrxiv.org\/lookup\/doi\/10.64898\/2026.08.10.26360107"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,12]]},"references-count":27,"URL":"https:\/\/doi.org\/10.64898\/2026.08.10.26360107","relation":{},"subject":[],"published":{"date-parts":[[2026,8,12]]},"subtype":"preprint"}}