{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"institution":[{"name":"medRxiv"}],"indexed":{"date-parts":[[2026,9,19]],"date-time":"2026-09-19T15:04:15Z","timestamp":1789830255317,"version":"4.0.1"},"posted":{"date-parts":[[2026,9,15]]},"group-title":"Health Informatics","reference-count":31,"publisher":"openRxiv","license":[{"start":{"date-parts":[[2026,9,15]],"date-time":"2026-09-15T00:00:00Z","timestamp":1789430400000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"}],"funder":[{"award":["R01HL168940"],"award-info":[{"award-number":["R01HL168940"]}],"id":[{"id":"https:\/\/ror.org\/001d55x84","id-type":"ROR","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"accepted":{"date-parts":[[2026,9,15]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                <jats:p>The limited availability of imaging data for uncommon cardiovascular phenotypes constrains the development of robust imaging models. We evaluated whether class-conditional diffusion models can generate synthetic transthoracic echocardiograms that improve downstream cardiac imaging tasks. The primary application was cardiac amyloidosis detection in a Duke University cohort using a two-step classifier, with external analyses using EchoNet-Dynamic for image-fidelity assessment and EchoNet-LVH for wall-thickness phenotype classification. The Duke cohort was partitioned at the patient-encounter level into 70% training, 15% validation, and 15% test sets; generators were trained only on the training partition, augmentation levels were selected using validation AUROC, and final evaluation used held-out real test data. In the Duke all-view two-step analysis, adding synthetic images increased AUROC from 0.883 to 0.924, with an AUROC difference of 0.041 (95% CI, 0.013\u20130.069); in EchoNet-LVH, AUROC increased from 0.832 to 0.864, with an AUROC difference of 0.032 (95% CI, 0.021\u20130.044). Expert review found that synthetic images were sometimes difficult to identify as synthetic, but rated them lower for diagnostic adequacy. These findings suggest that diffusion-generated echocardiograms may provide a practical approach to augmenting limited training data for selected cardiac imaging tasks and motivate further evaluation across clinical settings.<\/jats:p>","DOI":"10.64898\/2026.09.14.26363024","type":"posted-content","created":{"date-parts":[[2026,9,16]],"date-time":"2026-09-16T00:50:10Z","timestamp":1789519810000},"source":"Crossref","is-referenced-by-count":0,"title":["Synthetic Echocardiograms from Diffusion Models in Rare Cardiovascular Disease"],"prefix":"10.64898","author":[{"given":"Pengfei","family":"Lyu","sequence":"first","affiliation":[{"name":"Department of Biostatistics & Bioinformatics, Duke University, Durham, NC, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ricardo","family":"Henao","sequence":"additional","affiliation":[{"name":"Department of Biostatistics & Bioinformatics, Duke University, Durham, NC, USA"},{"name":"Duke Clinical Research Institute, Duke University School of Medicine, Durham, NC, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lydia Coulter","family":"Kwee","sequence":"additional","affiliation":[{"name":"Duke Health Center for Public Health, Duke University School of Medicine, Durham, NC, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fred Zhangzhi","family":"Peng","sequence":"additional","affiliation":[{"name":"Department of Biostatistics & Bioinformatics, Duke University, Durham, NC, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Melissa","family":"Hurdle","sequence":"additional","affiliation":[{"name":"Duke Health Center for Public Health, Duke University School of Medicine, Durham, NC, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sreekanth","family":"Vemulapalli","sequence":"additional","affiliation":[{"name":"Division of Cardiology, Department of Medicine, Duke University School of Medicine, Duke University, Durham, NC, USA"},{"name":"Duke Clinical Research Institute, Duke University School of Medicine, Durham, NC, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4980-845X","authenticated-orcid":false,"given":"Svati H.","family":"Shah","sequence":"additional","affiliation":[{"name":"Duke Health Center for Public Health, Duke University School of Medicine, Durham, NC, USA"},{"name":"Division of Cardiology, Department of Medicine, Duke University School of Medicine, Duke University, Durham, NC, USA"},{"name":"Duke Clinical Research Institute, Duke University School of Medicine, Durham, NC, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Michel Georges","family":"Khouri","sequence":"additional","affiliation":[{"name":"Division of Cardiology, Department of Medicine, Duke University School of Medicine, Duke University, Durham, NC, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8721-5252","authenticated-orcid":false,"given":"Anru","family":"Zhang","sequence":"additional","affiliation":[{"name":"Department of Biostatistics & Bioinformatics, Duke University, Durham, NC, USA"},{"name":"Department of Computer Science, Duke University, Durham, NC, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"54368","reference":[{"key":"2026091907300415000_2026.09.14.26363024v1.1","unstructured":"A. Raventan . Echo view classifier: Automated echocardiographic view classification. https:\/\/github.com\/raventan95\/echo-view-classifier, 2024. Accessed: 2024."},{"key":"2026091907300415000_2026.09.14.26363024v1.2","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-020-2145-8"},{"key":"2026091907300415000_2026.09.14.26363024v1.3","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-021-22877-8"},{"issue":"5","key":"2026091907300415000_2026.09.14.26363024v1.4","first-page":"101746","article-title":"Machine learning computer vision point of care decision support of echocardiographic identification of hypertrophic cardiomyopathy","volume":"4","year":"2025","journal-title":"JACC: Advances"},{"key":"2026091907300415000_2026.09.14.26363024v1.5","doi-asserted-by":"publisher","DOI":"10.1056\/NEJMra1814259"},{"issue":"1","key":"2026091907300415000_2026.09.14.26363024v1.6","doi-asserted-by":"crossref","first-page":"94","DOI":"10.1186\/s13023-020-01374-z","article-title":"Diagnosis support systems for rare diseases: a scoping review","volume":"15","year":"2020","journal-title":"Orphanet Journal of Rare Diseases"},{"issue":"9","key":"2026091907300415000_2026.09.14.26363024v1.7","doi-asserted-by":"crossref","first-page":"1263","DOI":"10.1109\/TKDE.2008.239","article-title":"Learning from imbalanced data","volume":"21","year":"2009","journal-title":"IEEE Transactions on Knowledge and Data Engineering"},{"key":"2026091907300415000_2026.09.14.26363024v1.8","doi-asserted-by":"publisher","DOI":"10.1016\/j.neunet.2018.07.011"},{"issue":"10","key":"2026091907300415000_2026.09.14.26363024v1.9","doi-asserted-by":"crossref","first-page":"273","DOI":"10.1007\/s10462-024-10884-2","article-title":"Handling imbalanced medical datasets: review of a decade of research","volume":"57","year":"2024","journal-title":"Artificial intelligence review"},{"key":"2026091907300415000_2026.09.14.26363024v1.10","doi-asserted-by":"publisher","DOI":"10.1007\/s10462-024-10759-6"},{"key":"2026091907300415000_2026.09.14.26363024v1.11","doi-asserted-by":"crossref","first-page":"321","DOI":"10.1613\/jair.953","article-title":"SMOTE: Synthetic minority over-sampling technique","volume":"16","year":"2002","journal-title":"Journal of Artificial Intelligence Research"},{"key":"2026091907300415000_2026.09.14.26363024v1.12","doi-asserted-by":"publisher","DOI":"10.1186\/1471-2105-14-106"},{"key":"2026091907300415000_2026.09.14.26363024v1.13","unstructured":"Charles Elkan . The foundations of cost-sensitive learning. In Proceedings of the Seventeenth International Joint Conference on Artificial Intelligence, 2001."},{"key":"2026091907300415000_2026.09.14.26363024v1.14","unstructured":"Pengfei Lyu , Zhengchi Ma , Linjun Zhang , and Anru R Zhang . Bias-corrected data synthesis for imbalanced learning. arXiv preprint arXiv:2510.26046, 2025."},{"key":"2026091907300415000_2026.09.14.26363024v1.15","unstructured":"Zhengchi Ma and Anru R Zhang . Synthetic augmentation in imbalanced learning: When it helps, when it hurts, and how much to add. arXiv preprint arXiv:2601.16120, 2026."},{"issue":"6","key":"2026091907300415000_2026.09.14.26363024v1.16","first-page":"2670","article-title":"Synthetic patient data generation and evaluation in disease prediction using small and imbalanced datasets","volume":"27","year":"2022","journal-title":"IEEE journal of biomedical and health informatics"},{"key":"2026091907300415000_2026.09.14.26363024v1.17","doi-asserted-by":"publisher","DOI":"10.1038\/s41591-024-02838-6"},{"key":"2026091907300415000_2026.09.14.26363024v1.18","doi-asserted-by":"publisher","DOI":"10.1038\/s41591-024-03359-y"},{"key":"2026091907300415000_2026.09.14.26363024v1.19","doi-asserted-by":"publisher","DOI":"10.1007\/s12350-019-01760-6"},{"key":"2026091907300415000_2026.09.14.26363024v1.20","unstructured":"Jascha Sohl-Dickstein , Eric Weiss , Niru Maheswaranathan , and Surya Ganguli . Deep unsupervised learning using nonequilibrium thermodynamics. In International conference on machine learning, pages 2256\u20132265. pmlr, 2015."},{"key":"2026091907300415000_2026.09.14.26363024v1.21","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.2006.11239"},{"key":"2026091907300415000_2026.09.14.26363024v1.22","unstructured":"Yang Song , Jascha Sohl-Dickstein , Diederik P Kingma , Abhishek Kumar , Stefano Ermon , and Ben Poole . Score-based generative modeling through stochastic differential equations. arXiv preprint arXiv:2011.13456, 2020."},{"key":"2026091907300415000_2026.09.14.26363024v1.23","unstructured":"Ian J Goodfellow , Jean Pouget-Abadie , Mehdi Mirza , Bing Xu , David Warde-Farley , Sherjil Ozair , Aaron Courville , and Yoshua Bengio . Generative adversarial nets. Advances in neural information processing systems, 27, 2014."},{"key":"2026091907300415000_2026.09.14.26363024v1.24","doi-asserted-by":"publisher","DOI":"10.1145\/3422622"},{"key":"2026091907300415000_2026.09.14.26363024v1.25","first-page":"8780","article-title":"Diffusion models beat gans on image synthesis","volume":"34","year":"2021","journal-title":"Advances in neural information processing systems"},{"key":"2026091907300415000_2026.09.14.26363024v1.26","doi-asserted-by":"publisher","DOI":"10.1093\/jamia\/ocae229"},{"issue":"4","key":"2026091907300415000_2026.09.14.26363024v1.27","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 cardiology"},{"key":"2026091907300415000_2026.09.14.26363024v1.28","unstructured":"Ozan Oktay , Jo Schlemper , Loic Le Folgoc , Matthew Lee , Mattias Heinrich , Kazunari Misawa , Kensaku Mori , Steven McDonagh , Nils Y Hammerla , Bernhard Kainz , et al. Attention u-net: Learning where to look for the pancreas. arXiv preprint arXiv:1804.03999, 2018."},{"key":"2026091907300415000_2026.09.14.26363024v1.29","unstructured":"Jonathan Ho and Tim Salimans . Classifier-free diffusion guidance. arXiv preprint arXiv:2207.12598, 2022."},{"key":"2026091907300415000_2026.09.14.26363024v1.30","unstructured":"Jiaming Song , Chenlin Meng , and Stefano Ermon . Denoising diffusion implicit models. arXiv preprint arXiv:2010.02502, 2020."},{"key":"2026091907300415000_2026.09.14.26363024v1.31","doi-asserted-by":"publisher","DOI":"10.2307\/2531595"}],"container-title":[],"original-title":[],"link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.64898\/2026.09.14.26363024","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,19]],"date-time":"2026-09-19T14:30:19Z","timestamp":1789828219000},"score":1,"resource":{"primary":{"URL":"http:\/\/medrxiv.org\/lookup\/doi\/10.64898\/2026.09.14.26363024"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,15]]},"references-count":31,"URL":"https:\/\/doi.org\/10.64898\/2026.09.14.26363024","relation":{},"subject":[],"published":{"date-parts":[[2026,9,15]]},"subtype":"preprint"}}