{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"institution":[{"name":"medRxiv"}],"indexed":{"date-parts":[[2026,1,16]],"date-time":"2026-01-16T13:25:22Z","timestamp":1768569922997,"version":"3.49.0"},"posted":{"date-parts":[[2025,4,30]]},"group-title":"Cardiovascular Medicine","reference-count":26,"publisher":"openRxiv","license":[{"start":{"date-parts":[[2025,4,30]],"date-time":"2025-04-30T00:00:00Z","timestamp":1745971200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"accepted":{"date-parts":[[2025,4,30]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                <jats:sec>\n                  <jats:title>Introduction<\/jats:title>\n                  <jats:p>Left ventricular diastolic dysfunction (LVDD) is most commonly evaluated by echocardiography. However, without a sole identifying metric, LVDD is assessed by a diagnostic algorithm relying on secondary characteristics that is laborious and has potential for interobserver variability.<\/jats:p>\n                <\/jats:sec>\n                <jats:sec>\n                  <jats:title>Methods<\/jats:title>\n                  <jats:p>To characterize concordance in clinical evaluations of LVDD, we evaluated historical echocardiogram studies at two academic medical centers for variability between clinician text reports and assessment by 2016 American Society of Echocardiography (ASE) guidelines. We then developed a workflow of 8 artificial intelligence (AI) models trained on over 155,000 studies to automate assessment of LVDD. Model performance was evaluated on temporally distinct held-out test sets from two academic medical centers.<\/jats:p>\n                <\/jats:sec>\n                <jats:sec>\n                  <jats:title>Results<\/jats:title>\n                  <jats:p>In a validation cohort of 955 studies from Cedars-Sinai Medical Center, our AI workflow demonstrated 76.5% agreement and a weighted Cohen\u2019s kappa of 0.52 with ASE guideline assessment using human measurements. In contrast, the clinician report evaluation had 48.5% agreement and a weighted Cohen\u2019s kappa of 0.29 with ASE guidelines. In the Stanford Healthcare cohort of 1,572 studies, the AI workflow had 66.7% agreement and a weighted Cohen\u2019s kappa of 0.27 with ASE guidelines, while the clinician assessment had 32.7% agreement and a weighted Cohen\u2019s kappa of 0.06. Performance was consistent across patient subgroups stratified by sex, age, hypertension, diabetes, obesity, and coronary artery disease.<\/jats:p>\n                <\/jats:sec>\n                <jats:sec>\n                  <jats:title>Conclusion<\/jats:title>\n                  <jats:p>Clinicians are often inconsistent in evaluating LVDD. We developed an AI pipeline that automates the clinical workflow of grading LVDD, which can contribute to improved diagnosis of heart failure.<\/jats:p>\n                <\/jats:sec>","DOI":"10.1101\/2025.04.29.25326683","type":"posted-content","created":{"date-parts":[[2025,4,30]],"date-time":"2025-04-30T23:30:21Z","timestamp":1746055821000},"source":"Crossref","is-referenced-by-count":0,"title":["Automated Deep Learning Pipeline for Characterizing Left Ventricular Diastolic Function"],"prefix":"10.64898","author":[{"given":"Victoria","family":"Yuan","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yuki","family":"Sahashi","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Hirotaka","family":"Ieki","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Milo\u0161","family":"Vukadinovic","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Christina","family":"Binder","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Konrad","family":"Pieszko","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Andrew P.","family":"Ambrosy","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Paul P.","family":"Cheng","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Susan","family":"Cheng","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"David","family":"Ouyang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"54368","reference":[{"key":"2025050510000823000_2025.04.29.25326683v2.1","doi-asserted-by":"crossref","first-page":"395","DOI":"10.1016\/j.echo.2021.12.003","article-title":"Associations between myocardial diastolic dysfunction and cardiovascular mortality in chronic kidney disease: A large single-center cohort study","volume":"35","year":"2022","journal-title":"J. Am. Soc. Echocardiogr"},{"key":"2025050510000823000_2025.04.29.25326683v2.2","doi-asserted-by":"crossref","first-page":"1925","DOI":"10.1093\/ckj\/sfad177","article-title":"Diastolic function in chronic kidney disease","volume":"16","year":"2023","journal-title":"Clin. Kidney J"},{"key":"2025050510000823000_2025.04.29.25326683v2.3","doi-asserted-by":"publisher","DOI":"10.1016\/j.jcmg.2018.10.038"},{"key":"2025050510000823000_2025.04.29.25326683v2.4","doi-asserted-by":"crossref","first-page":"1908","DOI":"10.1111\/echo.14698","article-title":"Diastolic function assessment by echocardiography: A practical manual for clinical use and future applications","volume":"37","year":"2020","journal-title":"Echocardiography"},{"key":"2025050510000823000_2025.04.29.25326683v2.5","doi-asserted-by":"crossref","first-page":"603247","DOI":"10.3389\/fphys.2020.603247","article-title":"The impact of antidiabetic therapies on diastolic dysfunction and diabetic cardiomyopathy","volume":"11","year":"2020","journal-title":"Front. Physiol"},{"key":"2025050510000823000_2025.04.29.25326683v2.6","doi-asserted-by":"crossref","first-page":"42","DOI":"10.1016\/S1062-1458(01)00250-1","article-title":"Pseudonormal and restrictive filling patterns predict left ventricular dilation and cardiac death after a first myocardial infarction: a serial color M-mode doppler echocardiographic study","volume":"10","year":"2001","journal-title":"ACC Curr. J. Rev"},{"key":"2025050510000823000_2025.04.29.25326683v2.7","doi-asserted-by":"publisher","DOI":"10.1161\/CIRCULATIONAHA.119.041818"},{"key":"2025050510000823000_2025.04.29.25326683v2.8","doi-asserted-by":"publisher","DOI":"10.1016\/j.jcmg.2018.10.039"},{"key":"2025050510000823000_2025.04.29.25326683v2.9","doi-asserted-by":"publisher","DOI":"10.1016\/j.echo.2016.01.011"},{"key":"2025050510000823000_2025.04.29.25326683v2.10","doi-asserted-by":"crossref","first-page":"1098","DOI":"10.1093\/ehjci\/jeab087","article-title":"EACVI survey on the evaluation of left ventricular diastolic function","volume":"22","year":"2021","journal-title":"Eur. Heart J. Cardiovasc. Imaging"},{"key":"2025050510000823000_2025.04.29.25326683v2.11","unstructured":"Clinical Course, and 1-Year Outcome in Pts Hospitalized for Heart Failure w Preserved Ejection Fraction (from the Polish Cohort of the ESC HF Long-Term Registry). Diagnosis."},{"key":"2025050510000823000_2025.04.29.25326683v2.12","doi-asserted-by":"publisher","DOI":"10.1093\/ehjci\/jeaa253"},{"key":"2025050510000823000_2025.04.29.25326683v2.13","doi-asserted-by":"crossref","first-page":"562","DOI":"10.1016\/j.echo.2024.01.001","article-title":"Precision of echocardiographic measurements","volume":"37","year":"2024","journal-title":"J. Am. Soc. Echocardiogr"},{"key":"2025050510000823000_2025.04.29.25326683v2.14","doi-asserted-by":"publisher","DOI":"10.1016\/j.jcmg.2011.01.016"},{"key":"2025050510000823000_2025.04.29.25326683v2.15","doi-asserted-by":"crossref","first-page":"925","DOI":"10.1016\/j.cardfail.2021.05.014","article-title":"Misdiagnosis of heart failure: A systematic review of the literature","volume":"27","year":"2021","journal-title":"J. Card. Fail"},{"key":"2025050510000823000_2025.04.29.25326683v2.16","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-020-2145-8"},{"key":"2025050510000823000_2025.04.29.25326683v2.17","doi-asserted-by":"crossref","first-page":"923","DOI":"10.1161\/CIRCULATIONAHA.124.069047","article-title":"High-throughput deep learning detection of mitral regurgitation. 1","volume":"150","year":"2024","journal-title":"Circulation"},{"key":"2025050510000823000_2025.04.29.25326683v2.18","doi-asserted-by":"crossref","first-page":"386","DOI":"10.1001\/jamacardio.2021.6059","article-title":"High-Throughput Precision Phenotyping of Left Ventricular Hypertrophy With Cardiovascular Deep Learning","volume":"7","year":"2022","journal-title":"JAMA Cardiol"},{"key":"2025050510000823000_2025.04.29.25326683v2.19","doi-asserted-by":"crossref","unstructured":"Sahashi, Y. et al. Artificial intelligence automation of echocardiographic measurements. Cardiovascular Medicine (2025).","DOI":"10.1101\/2025.03.18.25324215"},{"key":"2025050510000823000_2025.04.29.25326683v2.20","doi-asserted-by":"publisher","DOI":"10.1016\/S0735-1097(19)31924-2"},{"key":"2025050510000823000_2025.04.29.25326683v2.21","doi-asserted-by":"crossref","first-page":"10","DOI":"10.1038\/s41746-019-0216-8","article-title":"Deep learning interpretation of echocardiograms","volume":"3","year":"2020","journal-title":"NPJ Digit Med"},{"key":"2025050510000823000_2025.04.29.25326683v2.22","doi-asserted-by":"crossref","first-page":"1887","DOI":"10.1016\/j.jcmg.2021.04.010","article-title":"Deep-Learning Models for the Echocardiographic Assessment of Diastolic Dysfunction","volume":"14","year":"2021","journal-title":"JACC Cardiovasc. Imaging"},{"key":"2025050510000823000_2025.04.29.25326683v2.23","doi-asserted-by":"crossref","first-page":"4","DOI":"10.1038\/s41746-023-00993-7","article-title":"Artificial intelligence-enabled ECG for left ventricular diastolic function and filling pressure","volume":"7","year":"2024","journal-title":"NPJ Digit. Med"},{"key":"2025050510000823000_2025.04.29.25326683v2.24","doi-asserted-by":"crossref","first-page":"352","DOI":"10.21037\/cdt-24-25","article-title":"Artificial intelligence-enhanced automation of left ventricular diastolic assessment: a pilot study for feasibility, diagnostic validation, and outcome prediction","volume":"14","year":"2024","journal-title":"Cardiovasc. Diagn. Ther"},{"key":"2025050510000823000_2025.04.29.25326683v2.25","doi-asserted-by":"crossref","first-page":"6776","DOI":"10.1038\/s41467-022-34245-1","article-title":"A formal validation of a deep learning-based automated workflow for the interpretation of the echocardiogram","volume":"13","year":"2022","journal-title":"Nat. Commun"},{"key":"2025050510000823000_2025.04.29.25326683v2.26","unstructured":"Vukadinovic, M. et al. EchoPrime: A multi-video view-informed vision-language model for comprehensive echocardiography interpretation. arXiv [cs.CV] (2024)."}],"container-title":[],"original-title":[],"link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.1101\/2025.04.29.25326683","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,1,15]],"date-time":"2026-01-15T20:34:57Z","timestamp":1768509297000},"score":1,"resource":{"primary":{"URL":"http:\/\/medrxiv.org\/lookup\/doi\/10.1101\/2025.04.29.25326683"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,4,30]]},"references-count":26,"URL":"https:\/\/doi.org\/10.1101\/2025.04.29.25326683","relation":{},"subject":[],"published":{"date-parts":[[2025,4,30]]},"subtype":"preprint"}}