{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,4]],"date-time":"2026-08-04T03:25:26Z","timestamp":1785813926275,"version":"3.56.0"},"reference-count":40,"publisher":"Public Library of Science (PLoS)","issue":"5","license":[{"start":{"date-parts":[[2026,5,6]],"date-time":"2026-05-06T00:00:00Z","timestamp":1778025600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/100000069","name":"National Institute of Arthritis and Musculoskeletal and Skin Diseases","doi-asserted-by":"publisher","award":["R01-AR074453"],"award-info":[{"award-number":["R01-AR074453"]}],"id":[{"id":"10.13039\/100000069","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000069","name":"National Institute of Arthritis and Musculoskeletal and Skin Diseases","doi-asserted-by":"publisher","award":["N\/A (Salary)"],"award-info":[{"award-number":["N\/A (Salary)"]}],"id":[{"id":"10.13039\/100000069","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:p>Accurate disease prognosis is essential for patient care but is often hindered by the scarcity of longitudinal data. This study explores deep learning training strategies that utilize large, accessible diagnostic datasets to pretrain models aimed at predicting future disease progression in knee osteoarthritis (OA), Alzheimer\u2019s disease (AD), and breast cancer (BC). While diagnostic pretraining improves prognostic task performance, naive fine-tuning for prognosis can cause \u2018catastrophic forgetting,\u2019 where the model\u2019s original diagnostic accuracy degrades, a significant patient safety concern in real-world settings. To address this, we propose a sequential learning strategy with experience replay. We used cohorts with knee radiographs, brain MRIs, and digital mammograms to predict 4-year structural worsening in OA, 2-year cognitive decline in AD, and 5-year cancer diagnosis in BC. Our results showed that diagnostic pretraining on larger datasets improved prognosis model performance compared to standard baselines, boosting both the Area Under the Receiver Operating Characteristic curve (AUROC) (e.g., Knee OA external: 0.770 vs 0.747; Breast Cancer: 0.874 vs 0.848) and the Area Under the Precision-Recall Curve (AUPRC) (e.g., Alzheimer\u2019s Disease: 0.752 vs 0.683). Additionally, a sequential learning approach with experience replay achieved prognostic performance comparable to dedicated single-task models (e.g., Breast Cancer AUROC 0.876 vs 0.874) while also preserving diagnostic ability. This method maintained high diagnostic accuracy (e.g., Breast Cancer Balanced Accuracy 50.4% vs 50.9% for a dedicated diagnostic model), unlike simpler multitask methods prone to catastrophic forgetting (e.g., 37.7%). Our findings show that leveraging large diagnostic datasets is a reliable and data-efficient way to enhance prognostic models while maintaining essential diagnostic skills.<\/jats:p>","DOI":"10.1371\/journal.pone.0344600","type":"journal-article","created":{"date-parts":[[2026,5,6]],"date-time":"2026-05-06T17:32:02Z","timestamp":1778088722000},"page":"e0344600","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":1,"title":["Robust disease prognosis via diagnostic knowledge preservation: A sequential learning approach"],"prefix":"10.1371","volume":"21","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-7926-9090","authenticated-orcid":true,"given":"Haresh Rengaraj","family":"Rajamohan","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-0744-4183","authenticated-orcid":true,"given":"Yanqi","family":"Xu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Weicheng","family":"Zhu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Richard","family":"Kijowski","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kyunghyun","family":"Cho","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Krzysztof J.","family":"Geras","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Narges","family":"Razavian","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8809-5945","authenticated-orcid":true,"given":"Cem M.","family":"Deniz","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,5,6]]},"reference":[{"issue":"7","key":"pone.0344600.ref001","doi-asserted-by":"crossref","first-page":"1323","DOI":"10.1136\/annrheumdis-2013-204763","article-title":"The global burden of hip and knee osteoarthritis: estimates from the global burden of disease 2010 study","volume":"73","author":"M Cross","year":"2014","journal-title":"Annals of the Rheumatic Diseases"},{"issue":"9","key":"pone.0344600.ref002","article-title":"Global, regional, and national burden of osteoarthritis, 1990-2020 and projections to 2050: A systematic analysis for the Global Burden of Disease Study 2021","volume":"5","author":"GBD 2021 Osteoarthritis Collaborators","year":"2023","journal-title":"Lancet Rheumatol"},{"key":"pone.0344600.ref003","first-page":"123","article-title":"Alzheimer\u2019s disease - Why we need early diagnosis","volume":"9","author":"J Rasmussen","year":"2019","journal-title":"Degener Neurol Neuromuscul Dis"},{"key":"pone.0344600.ref004","doi-asserted-by":"crossref","first-page":"15","DOI":"10.1016\/j.breast.2022.08.010","article-title":"Current and future burden of breast cancer: Global statistics for 2020 and 2040","volume":"66","author":"M Arnold","year":"2022","journal-title":"The Breast"},{"issue":"10125","key":"pone.0344600.ref005","doi-asserted-by":"crossref","first-page":"1023","DOI":"10.1016\/S0140-6736(17)33326-3","article-title":"Global surveillance of trends in cancer survival 2000\u201314 (CONCORD-3): Analysis of individual records for 37 513 025 patients diagnosed with one of 18 cancers from 322 population-based registries in 71 countries","volume":"391","author":"C Allemani","year":"2018","journal-title":"The Lancet"},{"issue":"11","key":"pone.0344600.ref006","doi-asserted-by":"crossref","first-page":"2278","DOI":"10.1109\/5.726791","article-title":"Gradient-based learning applied to document recognition","volume":"86","author":"Y Lecun","year":"1998","journal-title":"Proc IEEE"},{"issue":"1","key":"pone.0344600.ref007","doi-asserted-by":"crossref","first-page":"1727","DOI":"10.1038\/s41598-018-20132-7","article-title":"Automatic knee osteoarthritis diagnosis from plain radiographs: A deep learning-based approach","volume":"8","author":"A Tiulpin","year":"2018","journal-title":"Scientific Reports"},{"issue":"1","key":"pone.0344600.ref008","doi-asserted-by":"crossref","first-page":"17106","DOI":"10.1038\/s41598-022-20674-x","article-title":"Generalizable deep learning model for early Alzheimer\u2019s disease detection from structural MRIs","volume":"12","author":"S Liu","year":"2022","journal-title":"Scientific Reports"},{"key":"pone.0344600.ref009","first-page":"184","article-title":"On the design of convolutional neural networks for automatic detection of Alzheimer\u2019s disease","author":"S Liu","year":"2020","journal-title":"Machine learning for health workshop"},{"issue":"1","key":"pone.0344600.ref010","doi-asserted-by":"crossref","first-page":"5210","DOI":"10.1038\/s41598-024-56001-9","article-title":"Automated detection of Alzheimer\u2019s disease: A multi-modal approach with 3D MRI and amyloid PET","volume":"14","author":"G Castellano","year":"2024","journal-title":"Sci Rep"},{"key":"pone.0344600.ref011","doi-asserted-by":"crossref","first-page":"569","DOI":"10.1016\/j.neuroimage.2014.06.077","article-title":"Hierarchical feature representation and multimodal fusion with deep learning for AD\/MCI diagnosis","volume":"101","author":"Alzheimer\u2019s Disease Neuroimaging Initiative","year":"2014","journal-title":"NeuroImage"},{"key":"pone.0344600.ref012","author":"KJ Geras","year":"2017"},{"issue":"4","key":"pone.0344600.ref013","doi-asserted-by":"crossref","first-page":"1184","DOI":"10.1109\/TMI.2019.2945514","article-title":"Deep neural networks improve radiologists\u2019 performance in breast cancer screening","volume":"39","author":"N Wu","year":"2020","journal-title":"IEEE Trans Med Imaging"},{"issue":"3","key":"pone.0344600.ref014","doi-asserted-by":"crossref","first-page":"584","DOI":"10.1148\/radiol.2020192091","article-title":"Prediction of total knee replacement and diagnosis of osteoarthritis by using deep learning on knee radiographs: Data from the osteoarthritis initiative","volume":"296","author":"K Leung","year":"2020","journal-title":"Radiology"},{"issue":"1","key":"pone.0344600.ref015","doi-asserted-by":"crossref","first-page":"6371","DOI":"10.1038\/s41598-020-63395-9","article-title":"Deep learning predicts total knee replacement from magnetic resonance images","volume":"10","author":"AA Tolpadi","year":"2020","journal-title":"Sci Rep"},{"issue":"1","key":"pone.0344600.ref016","doi-asserted-by":"crossref","first-page":"20038","DOI":"10.1038\/s41598-019-56527-3","article-title":"Multimodal machine learning-based knee osteoarthritis progression prediction from plain radiographs and clinical data","volume":"9","author":"A Tiulpin","year":"2019","journal-title":"Sci Rep"},{"issue":"1","key":"pone.0344600.ref017","doi-asserted-by":"crossref","first-page":"262","DOI":"10.1186\/s13075-021-02634-4","article-title":"A deep learning method for predicting knee osteoarthritis radiographic progression from MRI","volume":"23","author":"J-B Schiratti","year":"2021","journal-title":"Arthritis Res Ther"},{"issue":"1","key":"pone.0344600.ref018","doi-asserted-by":"crossref","first-page":"6922","DOI":"10.1038\/s41598-023-33934-1","article-title":"Prediction of total knee replacement using deep learning analysis of knee MRI","volume":"13","author":"HR Rajamohan","year":"2023","journal-title":"Sci Rep"},{"issue":"1","key":"pone.0344600.ref019","doi-asserted-by":"crossref","first-page":"1952","DOI":"10.1038\/s41598-018-37769-z","article-title":"Predicting Alzheimer\u2019s disease progression using multi-modal deep learning approach","volume":"9","author":"for Alzheimer\u2019s Disease Neuroimaging Initiative","year":"2019","journal-title":"Sci Rep"},{"key":"pone.0344600.ref020","doi-asserted-by":"crossref","first-page":"276","DOI":"10.1016\/j.neuroimage.2019.01.031","article-title":"A parameter-efficient deep learning approach to predict conversion from mild cognitive impairment to Alzheimer\u2019s disease","volume":"189","author":"Alzheimer\u2019s Disease Neuroimaging Initiative","year":"2019","journal-title":"Neuroimage"},{"issue":"3","key":"pone.0344600.ref021","doi-asserted-by":"crossref","first-page":"841","DOI":"10.1109\/TCBB.2018.2806438","article-title":"A Multimodal deep neural network for human breast cancer prognosis prediction by integrating multi-dimensional data","volume":"16","author":"D Sun","year":"2019","journal-title":"IEEE\/ACM Trans Comput Biol Bioinform"},{"issue":"1","key":"pone.0344600.ref022","doi-asserted-by":"crossref","first-page":"227","DOI":"10.1186\/s13244-024-01810-9","article-title":"Explainable breast cancer molecular expression prediction using multi-task deep-learning based on 3D whole breast ultrasound","volume":"15","author":"Z Huang","year":"2024","journal-title":"Insights Imaging"},{"key":"pone.0344600.ref023","doi-asserted-by":"crossref","first-page":"100687","DOI":"10.1016\/j.ocarto.2025.100687","article-title":"Artificial intelligence in osteoarthritis research: summary of the 2025 OARSI pre-congress workshop","volume":"1","author":"MS Harkey","year":"2025","journal-title":"Osteoarthritis and Cartilage Open"},{"issue":"7956","key":"pone.0344600.ref024","doi-asserted-by":"crossref","first-page":"259","DOI":"10.1038\/s41586-023-05881-4","article-title":"Foundation models for generalist medical artificial intelligence","volume":"616","author":"M Moor","year":"2023","journal-title":"Nature"},{"issue":"1","key":"pone.0344600.ref025","doi-asserted-by":"crossref","DOI":"10.1002\/alz.14321","article-title":"Overview of Alzheimer\u2019s disease neuroimaging initiative and future clinical trials","volume":"21","author":"MW Weiner","year":"2025","journal-title":"Alzheimers Dement"},{"key":"pone.0344600.ref026","article-title":"Foundation models for radiology: fundamentals, applications, opportunities, challenges, risks, and prospects","author":"T Akinci D\u2019Antonoli","year":"2025","journal-title":"Diagn Interv Radiol"},{"issue":"8071","key":"pone.0344600.ref027","doi-asserted-by":"crossref","first-page":"488","DOI":"10.1038\/s41586-025-09079-8","article-title":"A fully open AI foundation model applied to chest radiography","volume":"643","author":"D Ma","year":"2025","journal-title":"Nature"},{"issue":"2","key":"pone.0344600.ref028","doi-asserted-by":"crossref","first-page":"123","DOI":"10.1080\/09540099550039318","article-title":"Catastrophic forgetting, rehearsal and pseudorehearsal","volume":"7","author":"A Robins","year":"1995","journal-title":"Connection Science"},{"key":"pone.0344600.ref029","article-title":"On tiny episodic memories in continual learning","author":"A Chaudhry","year":"2019"},{"issue":"4","key":"pone.0344600.ref030","doi-asserted-by":"crossref","first-page":"494","DOI":"10.1136\/ard.16.4.494","article-title":"Radiological assessment of osteo-arthrosis","volume":"16","author":"JH Kellgren","year":"1957","journal-title":"Ann Rheum Dis"},{"issue":"3","key":"pone.0344600.ref031","doi-asserted-by":"crossref","first-page":"171","DOI":"10.4082\/kjfm.17.0090","article-title":"The Prevalence of osteoarthritis and risk factors in the korean population: The Sixth Korea National Health and Nutrition Examination Survey (VI-1, 2013)","volume":"40","author":"S Lee","year":"2019","journal-title":"Korean J Fam Med"},{"key":"pone.0344600.ref032","unstructured":"Nevitt M, Felson D, Lester G. The osteoarthritis initiative. Protocol for the cohort study. 2006;1(2)."},{"issue":"8","key":"pone.0344600.ref033","doi-asserted-by":"crossref","first-page":"10","DOI":"10.1016\/j.pmrj.2013.04.014","article-title":"The Multicenter Osteoarthritis Study (MOST): Opportunities for Rehabilitation Research","volume":"5","author":"NA Segal","year":"2013","journal-title":"PM & R: The Journal of Injury, Function, and Rehabilitation"},{"key":"pone.0344600.ref034","article-title":"The NYU breast cancer screening dataset v1. 0","author":"N Wu","year":"2019","journal-title":"New York, NY, USA: New York Univ"},{"key":"pone.0344600.ref035","doi-asserted-by":"crossref","first-page":"770","DOI":"10.1109\/CVPR.2016.90","article-title":"Deep Residual Learning for Image Recognition","author":"K He","year":"2016","journal-title":"2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)"},{"key":"pone.0344600.ref036","first-page":"3","article-title":"CBAM: Convolutional Block Attention Module","author":"S Woo","year":"2018","journal-title":"Lecture Notes in Computer Science. Springer International Publishing"},{"key":"pone.0344600.ref037","article-title":"Adam: A method for stochastic optimization","author":"DP Kingma","year":"2014","journal-title":"arXiv preprint arXiv:1412.6980"},{"issue":"1","key":"pone.0344600.ref038","doi-asserted-by":"crossref","first-page":"32","DOI":"10.1002\/1097-0142(1950)3:1<32::AID-CNCR2820030106>3.0.CO;2-3","article-title":"Index for rating diagnostic tests","volume":"3","author":"WJ Youden","year":"1950","journal-title":"Cancer"},{"issue":"3","key":"pone.0344600.ref039","doi-asserted-by":"crossref","first-page":"837","DOI":"10.2307\/2531595","article-title":"Comparing the areas under two or more correlated receiver operating characteristic curves: A nonparametric approach","volume":"44","author":"ER DeLong","year":"1988","journal-title":"Biometrics"},{"issue":"9","key":"pone.0344600.ref040","doi-asserted-by":"crossref","first-page":"1773","DOI":"10.1038\/s41591-022-01981-2","article-title":"Multimodal biomedical AI","volume":"28","author":"JN Acosta","year":"2022","journal-title":"Nat Med"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0344600","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,6]],"date-time":"2026-05-06T17:32:14Z","timestamp":1778088734000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0344600"}},"subtitle":[],"editor":[{"given":"Edward","family":"Hoffer","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,5,6]]},"references-count":40,"journal-issue":{"issue":"5","published-online":{"date-parts":[[2026,5,6]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0344600","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,6]]}}}