{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,29]],"date-time":"2026-09-29T18:00:24Z","timestamp":1790704824585,"version":"4.1.0"},"reference-count":50,"publisher":"BMJ","funder":[{"DOI":"10.13039\/100000050","name":"National Heart, Lung, and Blood Institute","doi-asserted-by":"publisher","award":["R00HL157421"],"award-info":[{"award-number":["R00HL157421"]}],"id":[{"id":"10.13039\/100000050","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000050","name":"National Heart, Lung, and Blood Institute","doi-asserted-by":"publisher","award":["R01AG091005"],"award-info":[{"award-number":["R01AG091005"]}],"id":[{"id":"10.13039\/100000050","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000050","name":"National Heart, Lung, and Blood Institute","doi-asserted-by":"publisher","award":["R01AG097715"],"award-info":[{"award-number":["R01AG097715"]}],"id":[{"id":"10.13039\/100000050","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000050","name":"National Heart, Lung, and Blood Institute","doi-asserted-by":"publisher","award":["R01HL173487"],"award-info":[{"award-number":["R01HL173487"]}],"id":[{"id":"10.13039\/100000050","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000050","name":"National Heart, Lung, and Blood Institute","doi-asserted-by":"publisher","award":["R01HL173526"],"award-info":[{"award-number":["R01HL173526"]}],"id":[{"id":"10.13039\/100000050","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000050","name":"National Heart, Lung, and Blood Institute","doi-asserted-by":"publisher","award":["R01HL173866"],"award-info":[{"award-number":["R01HL173866"]}],"id":[{"id":"10.13039\/100000050","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000968","name":"American Heart Association","doi-asserted-by":"publisher","award":["25POST1357984"],"award-info":[{"award-number":["25POST1357984"]}],"id":[{"id":"10.13039\/100000968","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100009210","name":"Food and Drug Administration","doi-asserted-by":"publisher","award":["U01FD008704"],"award-info":[{"award-number":["U01FD008704"]}],"id":[{"id":"10.13039\/100009210","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["bmj.com"],"crossmark-restriction":true},"short-container-title":["Heart"],"accepted":{"date-parts":[[2026,9,6]]},"abstract":"<jats:p>Artificial intelligence (AI) is reshaping cardiovascular medicine, evolving from task-specific supervised models towards multitask and foundation models capable of broader, more scalable interpretation. Early AI systems achieved high performance in electrocardiographic interpretation, echocardiographic measurement and disease diagnosis but their dependence on predefined outputs and labelled datasets limits flexibility and confines them to individual tasks. Foundation models address these limitations by learning reusable representations from large-scale data, often through self-supervised or contrastive pretraining, and adapting them to multiple downstream tasks. Emerging models in electrocardiography, echocardiography, chest radiography, cardiac MR and cardiac CT demonstrate capabilities including disease classification, quantitative measurement, segmentation, image-text retrieval, report generation and risk prediction. Generalist and multimodal biomedical foundation models further suggest a future in which cardiovascular AI integrates signals, imaging, clinical text, laboratory data and medical history to support patient-level diagnosis and decision-making. This review synthesises the transition from task-specific AI to cardiovascular foundation models and outlines opportunities for multimodal human-centred implementation.<\/jats:p>","DOI":"10.1136\/heartjnl-2026-328835","type":"journal-article","created":{"date-parts":[[2026,9,29]],"date-time":"2026-09-29T17:20:39Z","timestamp":1790702439000},"page":"heartjnl-2026-328835","update-policy":"https:\/\/doi.org\/10.1136\/crossmarkpolicy","source":"Crossref","is-referenced-by-count":0,"title":["From task-specific AI to cardiovascular foundation models: a new era of multimodal interpretation"],"prefix":"10.1136","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-6252-6134","authenticated-orcid":false,"given":"Hirotaka","family":"Ieki","sequence":"first","affiliation":[{"name":"Department of Medicine, Division of Cardiovascular Medicine, Stanford University School of Medicine, Stanford, California, USA"},{"name":"Kaiser Permanente Division of Research, Pleasanton, California, USA"},{"name":"Stanford Cardiovascular Institute, Stanford University, Stanford, California, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Paul","family":"Cheng","sequence":"additional","affiliation":[{"name":"Department of Medicine, Division of Cardiovascular Medicine, Stanford University School of Medicine, Stanford, California, USA"},{"name":"Stanford Cardiovascular Institute, Stanford University, Stanford, California, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Andrew P","family":"Ambrosy","sequence":"additional","affiliation":[{"name":"Kaiser Permanente Division of Research, Pleasanton, California, USA"},{"name":"Kaiser Permanente San Francisco Medical Center, San Francisco, California, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Alan C","family":"Kwan","sequence":"additional","affiliation":[{"name":"Department of Cardiology, Smidt Heart Institute, Cedars-Sinai Medical Center, Los Angeles, California, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Bryan","family":"He","sequence":"additional","affiliation":[{"name":"Kaiser Permanente Division of Research, Pleasanton, California, USA"},{"name":"Stanford University Department of Computer Science, Stanford, California, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Themistocles L","family":"Assimes","sequence":"additional","affiliation":[{"name":"Stanford Cardiovascular Institute, Stanford University, Stanford, California, USA"},{"name":"Stanford University School of Medicine, Stanford, California, USA"},{"name":"VA Palo Alto Health Care System, Palo Alto, California, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"James Y","family":"Zou","sequence":"additional","affiliation":[{"name":"Stanford University Department of Electrical Engineering, Stanford, California, USA"},{"name":"Stanford University Department of Biomedical Data Science, Stanford, California, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"David","family":"Ouyang","sequence":"additional","affiliation":[{"name":"Kaiser Permanente Division of Research, Pleasanton, California, USA"},{"name":"Department of Cardiology, Kaiser Permanente Santa Clara Medical Center, Santa Clara, California, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"239","published-online":{"date-parts":[[2026,9,29]]},"reference":[{"key":"2026092910202577000_heartjnl-2026-328835v1.1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jacc.2020.11.010"},{"key":"2026092910202577000_heartjnl-2026-328835v1.2","doi-asserted-by":"publisher","DOI":"10.1038\/s41591-021-01614-0"},{"key":"2026092910202577000_heartjnl-2026-328835v1.3","doi-asserted-by":"publisher","DOI":"10.1038\/s41591-018-0268-3"},{"key":"2026092910202577000_heartjnl-2026-328835v1.4","doi-asserted-by":"publisher","DOI":"10.1038\/nature21056"},{"key":"2026092910202577000_heartjnl-2026-328835v1.5","doi-asserted-by":"crossref","first-page":"520","DOI":"10.1038\/s41586-023-05947-3","article-title":"Blinded, randomized trial of sonographer versus AI cardiac function assessment","volume":"616","author":"He","year":"2023","journal-title":"Nature"},{"key":"2026092910202577000_heartjnl-2026-328835v1.6","doi-asserted-by":"crossref","first-page":"e367","DOI":"10.1016\/S2589-7500(24)00047-5","article-title":"Randomised controlled trials evaluating artificial intelligence in clinical practice: a scoping review","volume":"6","author":"Han","year":"2024","journal-title":"Lancet Digit Health"},{"key":"2026092910202577000_heartjnl-2026-328835v1.7","doi-asserted-by":"crossref","first-page":"2472","DOI":"10.1016\/j.jacc.2024.03.400","article-title":"Artificial Intelligence for Cardiovascular Care-Part 1: Advances: JACC Review Topic of the Week","volume":"83","author":"Elias","year":"2024","journal-title":"J Am Coll Cardiol"},{"key":"2026092910202577000_heartjnl-2026-328835v1.8","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-020-2145-8"},{"key":"2026092910202577000_heartjnl-2026-328835v1.9","doi-asserted-by":"publisher","DOI":"10.1038\/s41591-018-0240-2"},{"key":"2026092910202577000_heartjnl-2026-328835v1.10","doi-asserted-by":"crossref","DOI":"10.1161\/CIRCEP.119.007284","article-title":"Age and Sex Estimation Using Artificial Intelligence From Standard 12-Lead ECGs","volume":"12","author":"Attia","year":"2019","journal-title":"Circ Arrhythm Electrophysiol"},{"key":"2026092910202577000_heartjnl-2026-328835v1.11","doi-asserted-by":"crossref","DOI":"10.1038\/s41746-025-02050-x","article-title":"Artificial intelligence prediction of age from echocardiography as a marker for cardiovascular disease","volume":"8","author":"Rawlani","year":"2025","journal-title":"NPJ Digit Med"},{"key":"2026092910202577000_heartjnl-2026-328835v1.12","doi-asserted-by":"crossref","first-page":"159","DOI":"10.1038\/s43856-022-00220-6","article-title":"Deep learning-based age estimation from chest X-rays indicates cardiovascular prognosis","volume":"2","author":"Ieki","year":"2022","journal-title":"Commun Med (Lond)"},{"key":"2026092910202577000_heartjnl-2026-328835v1.13","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-018-0579-z"},{"key":"2026092910202577000_heartjnl-2026-328835v1.14","doi-asserted-by":"publisher","DOI":"10.1038\/s41588-021-00962-4"},{"key":"2026092910202577000_heartjnl-2026-328835v1.15","doi-asserted-by":"publisher","DOI":"10.1038\/s41588-022-01090-3"},{"key":"2026092910202577000_heartjnl-2026-328835v1.16","doi-asserted-by":"crossref","first-page":"2174","DOI":"10.1093\/eurheartj\/ehaf1119","article-title":"Foundation models for electrocardiogram interpretation: clinical implications","volume":"47","author":"Nolin-Lapalme","year":"2026","journal-title":"Eur Heart J"},{"key":"2026092910202577000_heartjnl-2026-328835v1.17","doi-asserted-by":"publisher","DOI":"10.1001\/jama.2025.8731"},{"key":"2026092910202577000_heartjnl-2026-328835v1.18","unstructured":"Crawshaw M . Multi-task learning with deep neural networks: a survey. arXiv 2020. doi:10.48550\/arXiv.2009.09796"},{"key":"2026092910202577000_heartjnl-2026-328835v1.19","unstructured":"Bommasani R , Hudson DA , Adeli E , et al . On the opportunities and risks of foundation models. arXiv 2021. doi:10.48550\/arXiv.2108.07258"},{"key":"2026092910202577000_heartjnl-2026-328835v1.20","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-023-05881-4"},{"key":"2026092910202577000_heartjnl-2026-328835v1.21","unstructured":"Radford A , Kim JW , Hallacy C , et al . Learning transferable visual models from natural language supervision. arXiv 2021. doi:10.48550\/arXiv.2103.00020"},{"key":"2026092910202577000_heartjnl-2026-328835v1.22","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-026-72946-z"},{"key":"2026092910202577000_heartjnl-2026-328835v1.23","doi-asserted-by":"crossref","DOI":"10.1093\/jamiaopen\/ooaf122","article-title":"ECG-FM: an open electrocardiogram foundation model","volume":"8","author":"McKeen","year":"2025","journal-title":"JAMIA Open"},{"key":"2026092910202577000_heartjnl-2026-328835v1.24","doi-asserted-by":"crossref","unstructured":"Li J , Aguirre AD , Moura V , et al . An Electrocardiogram Foundation Model Built on over 10 Million Recordings. NEJM AI 2025;2. doi:10.1056\/aioa2401033","DOI":"10.1056\/AIoa2401033"},{"key":"2026092910202577000_heartjnl-2026-328835v1.25","doi-asserted-by":"publisher","DOI":"10.1038\/s41591-024-02959-y"},{"key":"2026092910202577000_heartjnl-2026-328835v1.26","doi-asserted-by":"crossref","first-page":"970","DOI":"10.1038\/s41586-025-09850-x","article-title":"Comprehensive echocardiogram evaluation with view primed vision language AI","volume":"650","author":"Vukadinovic","year":"2026","journal-title":"Nature"},{"key":"2026092910202577000_heartjnl-2026-328835v1.27","doi-asserted-by":"crossref","unstructured":"Ji Y , Lin D , Wang X , et al . A Generative Foundation Model for Chest Radiography. NEJM AI 2026;3. doi:10.1056\/AIoa2500799","DOI":"10.1056\/AIoa2500799"},{"key":"2026092910202577000_heartjnl-2026-328835v1.28","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-025-09079-8"},{"key":"2026092910202577000_heartjnl-2026-328835v1.29","doi-asserted-by":"crossref","first-page":"4787","DOI":"10.1109\/TMI.2025.3581907","article-title":"Chest X-Ray Foundation Model With Global and Local Representations Integration","volume":"44","author":"Yang","year":"2025","journal-title":"IEEE Trans Med Imaging"},{"key":"2026092910202577000_heartjnl-2026-328835v1.30","doi-asserted-by":"crossref","first-page":"1399","DOI":"10.1038\/s41551-022-00936-9","article-title":"Expert-level detection of pathologies from unannotated chest X-ray images via self-supervised learning","volume":"6","author":"Tiu","year":"2022","journal-title":"Nat Biomed Eng"},{"key":"2026092910202577000_heartjnl-2026-328835v1.31","doi-asserted-by":"crossref","DOI":"10.1038\/s41746-025-02032-z","article-title":"EVA-X: a foundation model for general chest x-ray analysis with self-supervised learning","volume":"8","author":"Yao","year":"2025","journal-title":"NPJ Digit Med"},{"key":"2026092910202577000_heartjnl-2026-328835v1.32","unstructured":"Xu S , Yang L , Kelly C , et al . ELIXR: towards a general purpose X-ray artificial intelligence system through alignment of large language models and radiology vision encoders. arXiv 2023. doi:10.48550\/arXiv.2308.01317"},{"key":"2026092910202577000_heartjnl-2026-328835v1.33","doi-asserted-by":"crossref","DOI":"10.1038\/s41467-026-73022-2","article-title":"Contrastive language image pretraining for a cardiac magnetic resonance image embedding with zero-shot capabilities","volume":"17","author":"Nakashima","year":"2026","journal-title":"Nat Commun"},{"key":"2026092910202577000_heartjnl-2026-328835v1.34","doi-asserted-by":"crossref","DOI":"10.1038\/s43856-026-01636-0","article-title":"Development and validation of a versatile foundation model for cine cardiac magnetic resonance image analysis","volume":"6","author":"Fu","year":"2026","journal-title":"Commun Med (Lond)"},{"key":"2026092910202577000_heartjnl-2026-328835v1.35","unstructured":"Hu Y , Zheng Y , Miao S , et al . Cardiac-CLIP: a vision-language foundation model for 3D cardiac CT images. arXiv 2025. doi:10.48550\/arXiv.2507.22024"},{"key":"2026092910202577000_heartjnl-2026-328835v1.36","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-024-44824-z"},{"key":"2026092910202577000_heartjnl-2026-328835v1.37","doi-asserted-by":"crossref","unstructured":"Kirillov A , Mintun E , Ravi N , et al . Segment anything. arXiv 2023. doi:10.48550\/arXiv.2304.02643","DOI":"10.1109\/ICCV51070.2023.00371"},{"key":"2026092910202577000_heartjnl-2026-328835v1.38","unstructured":"Ma J , Yang Z , Kim S , et al . MedSAM2: segment anything in 3D medical images and videos. arXiv 2025. doi:10.48550\/arXiv.2504.03600"},{"key":"2026092910202577000_heartjnl-2026-328835v1.39","unstructured":"Liu A , Xue R , Cao XR , et al . MedSAM3: delving into segment anything with medical concepts. arXiv 2025. doi:10.48550\/arXiv.2511.19046"},{"key":"2026092910202577000_heartjnl-2026-328835v1.40","doi-asserted-by":"crossref","unstructured":"Tu T , Azizi S , Driess D , et al . Towards Generalist Biomedical AI. NEJM AI 2024;1. doi:10.1056\/AIoa2300138","DOI":"10.1056\/AIoa2300138"},{"key":"2026092910202577000_heartjnl-2026-328835v1.41","doi-asserted-by":"crossref","DOI":"10.1038\/s41746-021-00524-2","article-title":"Clinical impact and quality of randomized controlled trials involving interventions evaluating artificial intelligence prediction tools: a systematic review","volume":"4","author":"Zhou","year":"2021","journal-title":"NPJ Digit Med"},{"key":"2026092910202577000_heartjnl-2026-328835v1.42","doi-asserted-by":"crossref","DOI":"10.1136\/bmj-2023-078378","article-title":"TRIPOD+AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods","volume":"385","author":"Collins","year":"2024","journal-title":"BMJ"},{"key":"2026092910202577000_heartjnl-2026-328835v1.43","doi-asserted-by":"crossref","first-page":"4049","DOI":"10.1109\/TMI.2025.3580713","article-title":"EchoFM: Foundation Model for Generalizable Echocardiogram Analysis","volume":"44","author":"Kim","year":"2025","journal-title":"IEEE Trans Med Imaging"},{"key":"2026092910202577000_heartjnl-2026-328835v1.44","doi-asserted-by":"crossref","first-page":"19","DOI":"10.1038\/s41467-025-66340-4","article-title":"Echo-Vision-FM: a pre-training and fine-tuning framework for echocardiogram video vision foundation model","volume":"17","author":"Zhang","year":"2025","journal-title":"Nat Commun"},{"key":"2026092910202577000_heartjnl-2026-328835v1.45","unstructured":"Amadou AA , Zhang Y , Piat S , et al . EchoApex: a general-purpose vision foundation model for echocardiography. arXiv 2024. doi:10.48550\/arXiv.2410.11092"},{"key":"2026092910202577000_heartjnl-2026-328835v1.46","doi-asserted-by":"crossref","unstructured":"Joseph C , Mrudang M , Dhamanpreet K , et al . A multi-task deep learning model for pediatric echocardiography analysis. medRxiv 2025:2025.10.27.25338912. doi:10.1101\/2025.10.27.25338912","DOI":"10.1101\/2025.10.27.25338912"},{"key":"2026092910202577000_heartjnl-2026-328835v1.47","doi-asserted-by":"crossref","unstructured":"Yang Y , Peracchio L , Mayourian J , et al . An ECG foundation model for generalizable cardiac function prediction across the lifespan. medRxiv 2026:2026.05.26.26354128. doi:10.64898\/2026.05.26.26354128","DOI":"10.2139\/ssrn.6786041"},{"key":"2026092910202577000_heartjnl-2026-328835v1.48","doi-asserted-by":"crossref","DOI":"10.1093\/ehjdh\/ztag049","article-title":"Wearable-Echo-FM: an ECG echo foundation model for 1-lead electrocardiography","volume":"7","author":"Knight","year":"2026","journal-title":"Eur Heart J Digit Health"},{"key":"2026092910202577000_heartjnl-2026-328835v1.49","doi-asserted-by":"crossref","unstructured":"Shad R , Zakka C , Kaur D , et al . A generalizable deep learning system for cardiac MRI. Nat Biomed Eng 2026;1\u201316. doi:10.1038\/s41551-026-01637-3","DOI":"10.1038\/s41551-026-01637-3"},{"key":"2026092910202577000_heartjnl-2026-328835v1.50","unstructured":"Hao J , Durak G , Aktas HE , et al . CORA: a pathology synthesis driven foundation model for coronary CT angiography analysis and MACE risk assessment. arXiv 2026. doi:10.48550\/arXiv.2603.24847"}],"container-title":["Heart"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.1136\/heartjnl-2026-328835","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,29]],"date-time":"2026-09-29T17:20:46Z","timestamp":1790702446000},"score":1,"resource":{"primary":{"URL":"https:\/\/heart.bmj.com\/lookup\/doi\/10.1136\/heartjnl-2026-328835"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,29]]},"references-count":50,"alternative-id":["10.1136\/heartjnl-2026-328835"],"URL":"https:\/\/doi.org\/10.1136\/heartjnl-2026-328835","relation":{},"ISSN":["1355-6037","1468-201X"],"issn-type":[{"value":"1355-6037","type":"print"},{"value":"1468-201X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,29]]}}}