Show simple item record

dc.rights.licenseopenen_US
hal.structure.identifierBordeaux population health [BPH]
dc.contributor.authorDEVAUX, Anthony
hal.structure.identifierBordeaux population health [BPH]
dc.contributor.authorHELMER, Catherine
hal.structure.identifierStatistics In System biology and Translational Medicine [SISTM]
hal.structure.identifierBordeaux population health [BPH]
dc.contributor.authorGENUER, Robin
hal.structure.identifierBordeaux population health [BPH]
dc.contributor.authorPROUST LIMA, Cecile
ORCID: 0000-0002-9884-955X
IDREF: 114375747
dc.date.accessioned2023-12-13T15:21:46Z
dc.date.available2023-12-13T15:21:46Z
dc.date.issued2023-12-01
dc.identifier.issn1477-0334en_US
dc.identifier.urihttps://oskar-bordeaux.fr/handle/20.500.12278/186605
dc.description.abstractEnPredicting the individual risk of clinical events using the complete patient history is a major challenge in personalized medicine. Analytical methods have to account for a possibly large number of time-dependent predictors, which are often characterized by irregular and error-prone measurements, and are truncated early by the event. In this work, we extended the competing-risk random survival forests to handle such endogenous longitudinal predictors when predicting event probabilities. The method, implemented in the R package DynForest, internally transforms the time-dependent predictors at each node of each tree into time-fixed features (using mixed models) that can then be used as splitting candidates. The final individual event probability is computed as the average of leaf-specific Aalen-Johansen estimators over the trees. Using simulations, we compared the performances of DynForest to accurately predict an event with (i) a joint modeling alternative when considering two longitudinal predictors only, and with (ii) a regression calibration method that ignores the informative truncation by the event when dealing with a large number of longitudinal predictors. Through an application in dementia research, we also illustrated how DynForest can be used to develop a dynamic prediction tool for dementia from multimodal repeated markers, and quantify the importance of each marker.
dc.language.isoENen_US
dc.subject.enIndividual dynamic prediction
dc.subject.enMultivariate predictors
dc.subject.enRandom survival forest
dc.subject.enLongitudinal data
dc.subject.enSurvival data
dc.subject.enCompeting risks
dc.title.enRandom survival forests with multivariate longitudinal endogenous covariates
dc.title.alternativeStat Methods Med Resen_US
dc.typeArticle de revueen_US
dc.identifier.doi10.1177/09622802231206477en_US
dc.subject.halSciences du Vivant [q-bio]/Santé publique et épidémiologieen_US
dc.identifier.pubmed37886845en_US
bordeaux.journalStatistical Methods in Medical Researchen_US
bordeaux.page2331-2346en_US
bordeaux.volume32en_US
bordeaux.hal.laboratoriesBordeaux Population Health Research Center (BPH) - UMR 1219en_US
bordeaux.issue12en_US
bordeaux.institutionUniversité de Bordeauxen_US
bordeaux.institutionINSERMen_US
bordeaux.institutionINRIAen_US
bordeaux.teamLEHA_BPHen_US
bordeaux.teamSISTM_BPHen_US
bordeaux.teamBIOSTAT__BPHen_US
bordeaux.peerReviewedouien_US
bordeaux.inpressnonen_US
hal.popularnonen_US
hal.audienceInternationaleen_US
hal.exportfalse
dc.rights.ccPas de Licence CCen_US
bordeaux.COinSctx_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.jtitle=Statistical%20Methods%20in%20Medical%20Research&rft.date=2023-12-01&rft.volume=32&rft.issue=12&rft.spage=2331-2346&rft.epage=2331-2346&rft.eissn=1477-0334&rft.issn=1477-0334&rft.au=DEVAUX,%20Anthony&HELMER,%20Catherine&GENUER,%20Robin&PROUST%20LIMA,%20Cecile&rft.genre=article


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record