{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,2]],"date-time":"2026-04-02T05:36:43Z","timestamp":1775108203011,"version":"3.50.1"},"reference-count":38,"publisher":"Oxford University Press (OUP)","issue":"1","license":[{"start":{"date-parts":[[2024,11,21]],"date-time":"2024-11-21T00:00:00Z","timestamp":1732147200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Rhodes Trust and the EPSRC Centre for Doctoral Training in Health Data Science","award":["EP\/S02428X\/1"],"award-info":[{"award-number":["EP\/S02428X\/1"]}]},{"name":"Royal Academy of Engineering under the Research Fellowship scheme"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,1,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>Machine learning applications for longitudinal electronic health records often forecast the risk of events at fixed time points, whereas survival analysis achieves dynamic risk prediction by estimating time-to-event distributions. Here, we propose a novel conditional variational autoencoder-based method, DySurv, which uses a combination of static and longitudinal measurements from electronic health records to estimate the individual risk of death dynamically.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Materials and Methods<\/jats:title>\n                    <jats:p>DySurv directly estimates the cumulative risk incidence function without making any parametric assumptions on the underlying stochastic process of the time-to-event. We evaluate DySurv on 6 time-to-event benchmark datasets in healthcare, as well as 2 real-world intensive care unit (ICU) electronic health records (EHR) datasets extracted from the eICU Collaborative Research (eICU) and the Medical Information Mart for Intensive Care database (MIMIC-IV).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>DySurv outperforms other existing statistical and deep learning approaches to time-to-event analysis across concordance and other metrics. It achieves time-dependent concordance of over 60% in the eICU case. It is also over 12% more accurate and 22% more sensitive than in-use ICU scores like Acute Physiology and Chronic Health Evaluation (APACHE) and Sequential Organ Failure Assessment (SOFA) scores. The predictive capacity of DySurv is consistent and the survival estimates remain disentangled across different datasets.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>Our interdisciplinary framework successfully incorporates deep learning, survival analysis, and intensive care to create a novel method for time-to-event prediction from longitudinal health records. We test our method on several held-out test sets from a variety of healthcare datasets and compare it to existing in-use clinical risk scoring benchmarks.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>While our method leverages non-parametric extensions to deep learning-guided estimations of the survival distribution, further deep learning paradigms could be explored.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1093\/jamia\/ocae271","type":"journal-article","created":{"date-parts":[[2024,10,11]],"date-time":"2024-10-11T15:22:09Z","timestamp":1728660129000},"page":"112-122","source":"Crossref","is-referenced-by-count":8,"title":["DySurv: dynamic deep learning model for survival analysis with conditional variational inference"],"prefix":"10.1093","volume":"33","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-3757-7877","authenticated-orcid":false,"given":"Munib","family":"Mesinovic","sequence":"first","affiliation":[{"name":"Department of Engineering Science, University of Oxford , Oxford OX1 3PJ,","place":["United Kingdom"]}]},{"given":"Peter","family":"Watkinson","sequence":"additional","affiliation":[{"name":"Critical Care Research Group, Nuffield Department of Clinical Neurosciences, University of Oxford , Oxford OX3 7JX,","place":["United Kingdom"]}]},{"given":"Tingting","family":"Zhu","sequence":"additional","affiliation":[{"name":"Department of Engineering Science, University of Oxford , Oxford OX1 3PJ,","place":["United Kingdom"]}]}],"member":"286","published-online":{"date-parts":[[2024,11,21]]},"reference":[{"key":"2026010211513929400_ocae271-B1","doi-asserted-by":"crossref","first-page":"501","DOI":"10.1214\/088342306000000330","article-title":"Threshold regression for survival analysis: modeling event times by a stochastic process reaching a boundary","volume":"21","author":"Lee","year":"2006","journal-title":"Statist Sci"},{"key":"2026010211513929400_ocae271-B2","author":"Yoon","year":"2017"},{"key":"2026010211513929400_ocae271-B3","author":"Luck","year":"2017"},{"key":"2026010211513929400_ocae271-B4","first-page":"15111","article-title":"Deep extended hazard models for survival analysis","volume":"34","author":"Zhong","year":"2021","journal-title":"Advances in Neural Information Processing Systems"},{"key":"2026010211513929400_ocae271-B5","first-page":"1","author":"Kvamme","year":"2019"},{"key":"2026010211513929400_ocae271-B6","doi-asserted-by":"crossref","first-page":"926","DOI":"10.1093\/aje\/kwae002","article-title":"Why test for proportional hazards\u2014or any other model assumptions?","volume":"193","author":"Sj\u00f6lander","year":"2024","journal-title":"Am J Epidemiol"},{"key":"2026010211513929400_ocae271-B7","doi-asserted-by":"crossref","first-page":"24","DOI":"10.1186\/s12874-018-0482-1","article-title":"DeepSurv: personalized treatment recommender system using a Cox proportional hazards deep neural network","volume":"18","author":"Katzman","year":"2018","journal-title":"BMC Med Res Methodol"},{"key":"2026010211513929400_ocae271-B8","doi-asserted-by":"crossref","first-page":"65","DOI":"10.1007\/s10462-023-10681-3","article-title":"Deep learning for survival analysis: a review","volume":"57","author":"Wiegrebe","year":"2024","journal-title":"Artif Intell Rev"},{"key":"2026010211513929400_ocae271-B9","doi-asserted-by":"crossref","DOI":"10.1609\/aaai.v32i1.11842","article-title":"Deephit: a deep learning approach to survival analysis with competing risks","volume":"32","author":"Lee","year":"2018","journal-title":"AAAI"},{"key":"2026010211513929400_ocae271-B10","doi-asserted-by":"crossref","first-page":"122","DOI":"10.1109\/TBME.2019.2909027","article-title":"Dynamic-deephit: a deep learning approach for dynamic survival analysis with competing risks based on longitudinal data","volume":"67","author":"Lee","year":"2020","journal-title":"IEEE Trans Biomed Eng"},{"key":"2026010211513929400_ocae271-B11","author":"Giunchiglia","year":"2018"},{"key":"2026010211513929400_ocae271-B12","author":"Ren","year":"2019"},{"key":"2026010211513929400_ocae271-B13","doi-asserted-by":"crossref","first-page":"225","DOI":"10.1186\/s12911-020-01225-8","article-title":"Deep learning based feature-level integration of multi-omics data for breast cancer patients survival analysis","volume":"20","author":"Tong","year":"2020","journal-title":"BMC Med Inform Decis Mak"},{"key":"2026010211513929400_ocae271-B14","doi-asserted-by":"crossref","first-page":"i389","DOI":"10.1093\/bioinformatics\/btaa462","article-title":"Improved survival analysis by learning shared genomic information from pan-cancer data","volume":"36","author":"Kim","year":"2020","journal-title":"Bioinformatics"},{"key":"2026010211513929400_ocae271-B15","doi-asserted-by":"crossref","first-page":"339","DOI":"10.1016\/j.jcrc.2018.08.041","article-title":"Using survival analysis to predict septic shock onset in ICU patients","volume":"48","author":"Dummitt","year":"2018","journal-title":"J Crit Care"},{"key":"2026010211513929400_ocae271-B16","first-page":"10235","article-title":"Evidential sparsification of multimodal latent spaces in conditional variational autoencoders","volume":"33","author":"Itkina","year":"2020","journal-title":"Adv Neural Inf Processing Systems"},{"key":"2026010211513929400_ocae271-B17","doi-asserted-by":"crossref","first-page":"191","DOI":"10.7326\/0003-4819-122-3-199502010-00007","article-title":"The SUPPORT prognostic model: objective estimates of survival for seriously ill hospitalized adults","volume":"122","author":"Knaus","year":"1995","journal-title":"Ann Intern Med"},{"key":"2026010211513929400_ocae271-B18","doi-asserted-by":"crossref","first-page":"346","DOI":"10.1038\/nature10983","article-title":"The genomic and transcriptomic architecture of 2,000 breast tumours reveals novel subgroups","volume":"486","author":"Curtis","year":"2012","journal-title":"Nature"},{"key":"2026010211513929400_ocae271-B19","doi-asserted-by":"crossref","first-page":"2086","DOI":"10.1200\/JCO.1994.12.10.2086","article-title":"Randomized 2 x 2 trial evaluating hormonal treatment and the duration of chemotherapy in node-positive breast cancer patients. German Breast Cancer Study Group","volume":"12","author":"Schumacher","year":"1994","journal-title":"J Clin Oncol"},{"key":"2026010211513929400_ocae271-B20","doi-asserted-by":"crossref","first-page":"457","DOI":"10.1111\/1467-9876.00165","article-title":"Design and analysis of two-phase studies with binary outcome applied to Wilms tumour prognosis","volume":"48","author":"Breslow","year":"1999","journal-title":"J Roy Stat Soc Ser C Appl Stat"},{"key":"2026010211513929400_ocae271-B21","first-page":"710","author":"Kvamme","year":"2021"},{"key":"2026010211513929400_ocae271-B22","first-page":"1","author":"Kvamme","year":"2023"},{"key":"2026010211513929400_ocae271-B23","author":"Johnson","year":"2020"},{"key":"2026010211513929400_ocae271-B24","doi-asserted-by":"crossref","first-page":"180178","DOI":"10.1038\/sdata.2018.178","article-title":"The eICU Collaborative Research Database, a freely available multi-center database for critical care research","volume":"5","author":"Pollard","year":"2018","journal-title":"Sci Data"},{"key":"2026010211513929400_ocae271-B25","author":"Rocheteau","year":"2021"},{"key":"2026010211513929400_ocae271-B26","author":"Mesinovic","year":"2023"},{"key":"2026010211513929400_ocae271-B27","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3466782","article-title":"Attention-based deep recurrent model for survival prediction","volume":"2","author":"Sun","year":"2021","journal-title":"ACM Trans Comput Healthcare"},{"key":"2026010211513929400_ocae271-B28","first-page":"809","article-title":"Joint modeling of longitudinal and time-to-event data: an overview","author":"Tsiatis","year":"2004","journal-title":"Statistica Sinica"},{"key":"2026010211513929400_ocae271-B29","doi-asserted-by":"crossref","first-page":"431","DOI":"10.1038\/sj.bjc.6601119","article-title":"Survival analysis part II: multivariate data analysis\u2013an introduction to concepts and methods","volume":"89","author":"Bradburn","year":"2003","journal-title":"Br J Cancer"},{"key":"2026010211513929400_ocae271-B30","author":"Zhang","year":"2016"},{"key":"2026010211513929400_ocae271-B31","author":"Fil","year":"2021"},{"key":"2026010211513929400_ocae271-B32","author":"Kingma","year":"2013"},{"key":"2026010211513929400_ocae271-B33","first-page":"1","article-title":"Learning patient-specific cancer survival distributions as a sequence of dependent regressors","volume":"24","author":"Yu","year":"2011","journal-title":"Adv Neu Inf Process Syst"},{"key":"2026010211513929400_ocae271-B34","doi-asserted-by":"crossref","first-page":"e6257","DOI":"10.7717\/peerj.6257","article-title":"A scalable discrete-time survival model for neural networks","volume":"7","author":"Gensheimer","year":"2019","journal-title":"PeerJ"},{"key":"2026010211513929400_ocae271-B35","doi-asserted-by":"crossref","first-page":"3927","DOI":"10.1002\/sim.2427","article-title":"A time-dependent discrimination index for survival data","volume":"24","author":"Antolini","year":"2005","journal-title":"Stat Med"},{"key":"2026010211513929400_ocae271-B36","author":"Fotso","year":"2018"},{"key":"2026010211513929400_ocae271-B37","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1186\/cc13947","article-title":"Comparison of risk prediction scoring systems for ward patients: a retrospective nested case-control study","volume":"18","author":"Yu","year":"2014","journal-title":"Crit Care"},{"key":"2026010211513929400_ocae271-B38","doi-asserted-by":"crossref","first-page":"1654","DOI":"10.1097\/01.CCM.0000133021.22188.35","article-title":"Survival of critically ill patients hospitalized in and out of intensive care units under paucity of intensive care unit beds","volume":"32","author":"Simchen","year":"2004","journal-title":"Crit Care Med"}],"container-title":["Journal of the American Medical Informatics Association"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/jamia\/article-pdf\/33\/1\/112\/60775529\/ocae271.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/jamia\/article-pdf\/33\/1\/112\/60775529\/ocae271.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,1,2]],"date-time":"2026-01-02T16:51:47Z","timestamp":1767372707000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/jamia\/article\/33\/1\/112\/7906103"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,11,21]]},"references-count":38,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2024,11,21]]},"published-print":{"date-parts":[[2026,1,1]]}},"URL":"https:\/\/doi.org\/10.1093\/jamia\/ocae271","relation":{},"ISSN":["1067-5027","1527-974X"],"issn-type":[{"value":"1067-5027","type":"print"},{"value":"1527-974X","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2026,1]]},"published":{"date-parts":[[2024,11,21]]}}}