{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,11,12]],"date-time":"2026-11-12T10:31:42Z","timestamp":1794479502080,"version":"4.9.1"},"reference-count":23,"publisher":"Springer Science and Business Media LLC","issue":"11","license":[{"start":{"date-parts":[[2026,10,13]],"date-time":"2026-10-13T00:00:00Z","timestamp":1791849600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,10,13]],"date-time":"2026-10-13T00:00:00Z","timestamp":1791849600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100000156","name":"Fonds de Recherche du Qu\u00e9bec - Sant\u00e9","doi-asserted-by":"publisher","award":["https:\/\/doi.org\/10.69777\/380127"],"award-info":[{"award-number":["https:\/\/doi.org\/10.69777\/380127"]}],"id":[{"id":"10.13039\/501100000156","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Nat Cardiovasc Res"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>\n                    Invasive coronary angiography is the reference standard for diagnosing coronary artery disease, yet assessment is subjective, and existing artificial-intelligence tools typically assess stenosis in individual frames or videos and do not integrate the full multiview angiographic study. Here we show that DeepCORO-CLIP, trained on 203,808 angiographic videos and reports from 28,117 patients, interprets whole multiple-video angiograms at once. It detects obstructive stenoses (\u226570%), with an area under the receiver operating characteristic curve (AUROC) of 0.89 internally and externally, and identifies chronic total occlusion (AUROC\u2009=\u20090.90 to 0.95), intra-coronary thrombus and calcification (AUROC\u2009=\u20090.90 to 0.91). It also predicts 1-year cardiovascular events (AUROC\u2009=\u20090.79, 95% CI of 0.74 to 0.84, versus 0.65 for a model derived using clinical variables) and left ventricular ejection fraction (mean absolute error 7.3% versus 8.8% for a previous model), both achieved using coronary angiography alone. In a blinded review of 365 evaluations across 288 angiograms, its reports were clinically usable in 76% of evaluations versus 88% for cardiologists, with comparable false positives (17% versus 14%,\n                    <jats:italic>P<\/jats:italic>\n                    \u2009=\u20090.38); residual missed stenoses reflected an adjustable threshold rather than an absent signal.\n                  <\/jats:p>","DOI":"10.1038\/s44161-026-00890-8","type":"journal-article","created":{"date-parts":[[2026,10,13]],"date-time":"2026-10-13T19:38:07Z","timestamp":1791920287000},"page":"1179-1192","update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["A multiview video\u2013text model detects obstructive coronary stenosis and predicts 1-year cardiovascular events from angiography"],"prefix":"10.1038","volume":"5","author":[{"given":"Sarra","family":"Harrabi","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-2771-1505","authenticated-orcid":false,"given":"Yichen","family":"Wu","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-6370-5816","authenticated-orcid":false,"given":"Omar","family":"Majwal","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Gabriel","family":"Asslo","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-5663-6413","authenticated-orcid":false,"given":"George","family":"Azar","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0310-3326","authenticated-orcid":false,"given":"Geoffrey H.","family":"Tison","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Minhaj","family":"Ansari","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Milos","family":"Vukadinovic","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"David","family":"Ouyang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1618-6244","authenticated-orcid":false,"given":"Joshua P.","family":"Barrios","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0001-0466-1354","authenticated-orcid":false,"given":"Jacques","family":"Delfrate","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8490-0270","authenticated-orcid":false,"given":"Robert","family":"Avram","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"},{"role":"corresponding-author","vocabulary":"crossref"}]}],"member":"297","published-online":{"date-parts":[[2026,10,13]]},"reference":[{"key":"890_CR1","doi-asserted-by":"publisher","first-page":"e9","DOI":"10.1161\/CIR.0000000000001168","volume":"148","author":"SS Virani","year":"2023","unstructured":"Virani, S. S. et al. 2023 AHA\/ACC\/ACCP\/ASPC\/NLA\/PCNA guideline for the management of patients with chronic coronary disease: a report of the American Heart Association\/American College of Cardiology Joint Committee on Clinical Practice Guidelines. Circulation 148, e9\u2013e119 (2023).","journal-title":"Circulation"},{"key":"890_CR2","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1016\/j.ahj.2016.10.014","volume":"184","author":"R Shah","year":"2017","unstructured":"Shah, R. et al. Comparison of visual assessment of coronary stenosis with independent quantitative coronary angiography: findings from the Prospective Multicenter Imaging Study for Evaluation of Chest Pain (PROMISE) trial. Am. Heart J. 184, 1\u20139 (2017).","journal-title":"Am. Heart J."},{"key":"890_CR3","doi-asserted-by":"publisher","first-page":"106","DOI":"10.1016\/S0002-8703(00)90316-8","volume":"139","author":"LL Leape","year":"2000","unstructured":"Leape, L. L. et al. Effect of variability in the interpretation of coronary angiograms on the appropriateness of use of coronary revascularization procedures. Am. Heart J. 139, 106\u2013113 (2000).","journal-title":"Am. Heart J."},{"key":"890_CR4","doi-asserted-by":"publisher","first-page":"1793","DOI":"10.1161\/CIRCULATIONAHA.113.001952","volume":"127","author":"BK Nallamothu","year":"2013","unstructured":"Nallamothu, B. K. et al. Comparison of clinical interpretation with visual assessment and quantitative coronary angiography in patients undergoing percutaneous coronary intervention in contemporary practice. Circulation 127, 1793\u20131800 (2013).","journal-title":"Circulation"},{"key":"890_CR5","doi-asserted-by":"publisher","first-page":"142","DOI":"10.1038\/s41746-023-00880-1","volume":"6","author":"R Avram","year":"2023","unstructured":"Avram, R. et al. CathAI: fully automated coronary angiography interpretation and stenosis estimation. npj Digit. Med. 6, 142 (2023).","journal-title":"npj Digit. Med."},{"key":"890_CR6","doi-asserted-by":"publisher","first-page":"138","DOI":"10.1038\/s41746-024-01134-4","volume":"7","author":"\u00c9L Langlais","year":"2024","unstructured":"Langlais, \u00c9. L. et al. Evaluation of stenoses using AI video models applied to coronary angiography. npj Digit. Med. 7, 138 (2024).","journal-title":"npj Digit. Med."},{"key":"890_CR7","doi-asserted-by":"publisher","first-page":"970","DOI":"10.1038\/s41586-025-09850-x","volume":"650","author":"M Vukadinovic","year":"2026","unstructured":"Vukadinovic, M. et al. Comprehensive echocardiogram evaluation with view primed vision language AI. Nature 650, 970\u2013977 https:\/\/doi.org\/10.1038\/s41586-025-09850-x (2026).","journal-title":"Nature"},{"key":"890_CR8","doi-asserted-by":"publisher","first-page":"1481","DOI":"10.1038\/s41591-024-02959-y","volume":"30","author":"M Christensen","year":"2024","unstructured":"Christensen, M., Vukadinovic, M., Yuan, N. & Ouyang, D. Vision\u2013language foundation model for echocardiogram interpretation. Nat. Med. 30, 1481\u20131488 https:\/\/doi.org\/10.1038\/s41591-024-02959-y (2024).","journal-title":"Nat. Med."},{"key":"890_CR9","doi-asserted-by":"publisher","first-page":"586","DOI":"10.1001\/jamacardio.2023.0968","volume":"8","author":"R Avram","year":"2023","unstructured":"Avram, R. et al. Automated assessment of cardiac systolic function from coronary angiograms with video-based artificial intelligence algorithms. JAMA Cardiol. 8, 586\u2013594 (2023).","journal-title":"JAMA Cardiol."},{"key":"890_CR10","doi-asserted-by":"publisher","first-page":"1828","DOI":"10.1016\/j.cjca.2024.05.025","volume":"40","author":"P Theriault-Lauzier","year":"2024","unstructured":"Theriault-Lauzier, P. et al. A responsible framework for applying artificial intelligence on medical images and signals at the point-of-care: the PACS-AI platform. Can. J. Cardiol. 40, 1828\u20131840 https:\/\/doi.org\/10.1016\/j.cjca.2024.05.025 (2024).","journal-title":"Can. J. Cardiol."},{"key":"890_CR11","doi-asserted-by":"publisher","first-page":"2139","DOI":"10.1007\/s10554-019-01672-z","volume":"35","author":"A Hideo-Kajita","year":"2019","unstructured":"Hideo-Kajita, A. et al. Impact of two formulas to calculate percentage diameter stenosis of coronary lesions: from stenosis models (phantom lesion model) to actual clinical lesions. Int. J. Cardiovasc. Imaging 35, 2139\u20132146 (2019).","journal-title":"Int. J. Cardiovasc. Imaging"},{"key":"890_CR12","doi-asserted-by":"publisher","first-page":"32","DOI":"10.4244\/EIJ-D-20-00570","volume":"17","author":"T Du","year":"2021","unstructured":"Du, T. et al. Training and validation of a deep learning architecture for the automatic analysis of coronary angiography. Eurointervention 17, 32\u201340 (2021).","journal-title":"Eurointervention"},{"key":"890_CR13","unstructured":"Munim, A., et al. EchoJEPA: a latent predictive foundation model for echocardiography. Preprint at https:\/\/arxiv.org\/abs\/2602.02603 (2026)."},{"key":"890_CR14","doi-asserted-by":"publisher","first-page":"9","DOI":"10.1016\/j.ahj.2016.03.016","volume":"177","author":"G Ndrepepa","year":"2016","unstructured":"Ndrepepa, G. et al. Association of progression or regression of coronary artery atherosclerosis with long-term prognosis. Am. Hear. J. 177, 9\u201316 (2016).","journal-title":"Am. Hear. J."},{"key":"890_CR15","doi-asserted-by":"publisher","first-page":"199","DOI":"10.1016\/S0140-6736(20)32519-8","volume":"397","author":"F D\u2019Ascenzo","year":"2021","unstructured":"D\u2019Ascenzo, F. et al. Machine learning-based prediction of adverse events following an acute coronary syndrome (PRAISE): a modelling study of pooled datasets. Lancet 397, 199\u2013207 (2021).","journal-title":"Lancet"},{"key":"890_CR16","doi-asserted-by":"publisher","first-page":"1735","DOI":"10.1016\/j.jacc.2022.08.767","volume":"80","author":"G Batra","year":"2022","unstructured":"Batra, G. et al. Biomarker-based prediction of recurrent ischemic events in patients with acute coronary syndromes. J. Am. Coll. Cardiol. 80, 1735\u20131747 (2022).","journal-title":"J. Am. Coll. Cardiol."},{"key":"890_CR17","doi-asserted-by":"publisher","first-page":"e091028","DOI":"10.1136\/bmjopen-2024-091028","volume":"15","author":"R Kumar","year":"2025","unstructured":"Kumar, R. et al. Assessment of the prognostic performance of TIMI, PAMI, CADILLAC and GRACE scores for short-term major adverse cardiovascular events in patients undergoing emergent percutaneous revascularisation: a prospective observational study. BMJ Open 15, e091028 (2025).","journal-title":"BMJ Open"},{"key":"890_CR18","doi-asserted-by":"publisher","first-page":"1384","DOI":"10.1016\/j.cjca.2025.12.024","volume":"42","author":"E Bogner","year":"2026","unstructured":"Bogner, E. et al. Comprehensive machine learning-enabled outcome prediction for patients with coronary artery disease using multicentre patient data. Can. J. Cardiol. 42, 1384\u20131396 https:\/\/doi.org\/10.1016\/j.cjca.2025.12.024 (2026).","journal-title":"Can. J. Cardiol."},{"key":"890_CR19","doi-asserted-by":"publisher","first-page":"e078378","DOI":"10.1136\/bmj-2023-078378","volume":"385","author":"GS Collins","year":"2024","unstructured":"Collins, G. S. et al. TRIPOD+AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods. Brit. Med. J. 385, e078378 (2024).","journal-title":"Brit. Med. J."},{"key":"890_CR20","doi-asserted-by":"publisher","first-page":"968","DOI":"10.1136\/hrt.2005.063107","volume":"91","author":"CD Mario","year":"2005","unstructured":"Mario, C. D. & Sutaria, N. Coronary angiography in the angioplasty era: projections with a meaning. Heart 91, 968 (2005).","journal-title":"Heart"},{"key":"890_CR21","first-page":"219","volume":"1","author":"S Georgios","year":"2005","unstructured":"Georgios, S. et al. The SYNTAX Score: an angiographic tool grading the complexity of coronary artery disease. EuroIntervention 1, 219\u2013227 (2005).","journal-title":"EuroIntervention"},{"key":"890_CR22","unstructured":"DeepSeek-AI, et al. DeepSeek-V2: a strong, economical, and efficient mixture-of-experts language model. Preprint at https:\/\/arxiv.org\/abs\/2405.04434 (2024)."},{"key":"890_CR23","doi-asserted-by":"publisher","first-page":"837","DOI":"10.2307\/2531595","volume":"44","author":"ER DeLong","year":"1988","unstructured":"DeLong, E. R., DeLong, D. M. & Clarke-Pearson, D. L. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics 44, 837\u2013845 (1988).","journal-title":"Biometrics"}],"container-title":["Nature Cardiovascular Research"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s44161-026-00890-8.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s44161-026-00890-8","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s44161-026-00890-8.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,11,12]],"date-time":"2026-11-12T10:07:12Z","timestamp":1794478032000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s44161-026-00890-8"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,10,13]]},"references-count":23,"journal-issue":{"issue":"11","published-online":{"date-parts":[[2026,11]]}},"alternative-id":["890"],"URL":"https:\/\/doi.org\/10.1038\/s44161-026-00890-8","relation":{},"ISSN":["2731-0590"],"issn-type":[{"value":"2731-0590","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,10,13]]},"assertion":[{"value":"3 June 2026","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"11 September 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"13 October 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"R.A. and G.H.T. are, together with Jeffrey E. Olgin, named inventors on patent application WO 2022\/261641 A1 (PCT\/US2022\/072817; Method and System for Automated Analysis of Coronary Angiograms; priority date 8 June 2021). The applicant is The Regents of the University of California, an institution and not the authors; the corresponding US national-phase application 18\/568,693 was published as US 2024\/0273723 A1 and is pending. The application covers single-video deep-learning estimation of stenosis severity and left ventricular ejection fraction from coronary angiograms and does not cover any aspect of the multiview video\u2013text model reported in this paper. R.A. holds stock options in Spiralis; holds stock in Divoco AI; is a co-founder of and holds equity in FrontRx; has received speaker fees from Abbott, Boston Scientific, Boehringer Ingelheim and Novartis; and has received an investigator-initiated grant from Bristol Myers Squibb. G.H.T. has received research grants from Myokardia, General Electric and Janssen Pharmaceuticals. The other authors declare no competing interests.","order":1,"name":"Ethics","label":"Competing interests","group":{"name":"EthicsHeading","label":"Ethics"}}]}}