001005756 001__ 1005756
001005756 005__ 20230929112520.0
001005756 0247_ $$2doi$$a10.1038/s41598-023-30986-1
001005756 0247_ $$2Handle$$a2128/34225
001005756 0247_ $$2pmid$$a36906642
001005756 0247_ $$2WOS$$aWOS:000958974100039
001005756 037__ $$aFZJ-2023-01610
001005756 041__ $$aEnglish
001005756 082__ $$a600
001005756 1001_ $$0P:(DE-HGF)0$$aMerkelbach, Kilian$$b0
001005756 245__ $$aNovel architecture for gated recurrent unit autoencoder trained on time series from electronic health records enables detection of ICU patient subgroups
001005756 260__ $$a[London]$$bMacmillan Publishers Limited, part of Springer Nature$$c2023
001005756 3367_ $$2DRIVER$$aarticle
001005756 3367_ $$2DataCite$$aOutput Types/Journal article
001005756 3367_ $$0PUB:(DE-HGF)16$$2PUB:(DE-HGF)$$aJournal Article$$bjournal$$mjournal$$s1680244024_32318
001005756 3367_ $$2BibTeX$$aARTICLE
001005756 3367_ $$2ORCID$$aJOURNAL_ARTICLE
001005756 3367_ $$00$$2EndNote$$aJournal Article
001005756 520__ $$aElectronic health records (EHRs) are used in hospitals to store diagnoses, clinician notes, examinations, lab results, and interventions for each patient. Grouping patients into distinct subsets, for example, via clustering, may enable the discovery of unknown disease patterns or comorbidities, which could eventually lead to better treatment through personalized medicine. Patient data derived from EHRs is heterogeneous and temporally irregular. Therefore, traditional machine learning methods like PCA are ill-suited for analysis of EHR-derived patient data. We propose to address these issues with a new methodology based on training a gated recurrent unit (GRU) autoencoder directly on health record data. Our method learns a low-dimensional feature space by training on patient data time series, where the time of each data point is expressed explicitly. We use positional encodings for time, allowing our model to better handle the temporal irregularity of the data. We apply our method to data from the Medical Information Mart for Intensive Care (MIMIC-III). Using our data-derived feature space, we can cluster patients into groups representing major classes of disease patterns. Additionally, we show that our feature space exhibits a rich substructure at multiple scales.
001005756 536__ $$0G:(DE-HGF)POF4-5112$$a5112 - Cross-Domain Algorithms, Tools, Methods Labs (ATMLs) and Research Groups (POF4-511)$$cPOF4-511$$fPOF IV$$x0
001005756 536__ $$0G:(BMBF)01ZZ1803M$$aSMITH - Medizininformatik-Konsortium - Beitrag Forschungszentrum Jülich (01ZZ1803M)$$c01ZZ1803M$$x1
001005756 588__ $$aDataset connected to CrossRef, Journals: juser.fz-juelich.de
001005756 7001_ $$0P:(DE-HGF)0$$aSchaper, Steffen$$b1
001005756 7001_ $$0P:(DE-HGF)0$$aDiedrich, Christian$$b2
001005756 7001_ $$0P:(DE-Juel1)185651$$aFritsch, Sebastian Johannes$$b3$$ufzj
001005756 7001_ $$0P:(DE-HGF)0$$aSchuppert, Andreas$$b4$$eCorresponding author
001005756 773__ $$0PERI:(DE-600)2615211-3$$a10.1038/s41598-023-30986-1$$gVol. 13, no. 1, p. 4053$$n1$$p4053$$tScientific reports$$v13$$x2045-2322$$y2023
001005756 8564_ $$uhttps://juser.fz-juelich.de/record/1005756/files/s41598-023-30986-1.pdf$$yOpenAccess
001005756 909CO $$ooai:juser.fz-juelich.de:1005756$$pdnbdelivery$$pdriver$$pVDB$$popen_access$$popenaire
001005756 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)185651$$aForschungszentrum Jülich$$b3$$kFZJ
001005756 9131_ $$0G:(DE-HGF)POF4-511$$1G:(DE-HGF)POF4-510$$2G:(DE-HGF)POF4-500$$3G:(DE-HGF)POF4$$4G:(DE-HGF)POF$$9G:(DE-HGF)POF4-5112$$aDE-HGF$$bKey Technologies$$lEngineering Digital Futures – Supercomputing, Data Management and Information Security for Knowledge and Action$$vEnabling Computational- & Data-Intensive Science and Engineering$$x0
001005756 9141_ $$y2023
001005756 915__ $$0StatID:(DE-HGF)0510$$2StatID$$aOpenAccess
001005756 915__ $$0LIC:(DE-HGF)CCBY4$$2HGFVOC$$aCreative Commons Attribution CC BY 4.0
001005756 915__ $$0StatID:(DE-HGF)0100$$2StatID$$aJCR$$bSCI REP-UK : 2022$$d2023-08-24
001005756 915__ $$0StatID:(DE-HGF)0200$$2StatID$$aDBCoverage$$bSCOPUS$$d2023-08-24
001005756 915__ $$0StatID:(DE-HGF)0300$$2StatID$$aDBCoverage$$bMedline$$d2023-08-24
001005756 915__ $$0StatID:(DE-HGF)0320$$2StatID$$aDBCoverage$$bPubMed Central$$d2023-08-24
001005756 915__ $$0StatID:(DE-HGF)0501$$2StatID$$aDBCoverage$$bDOAJ Seal$$d2023-04-12T15:11:06Z
001005756 915__ $$0StatID:(DE-HGF)0500$$2StatID$$aDBCoverage$$bDOAJ$$d2023-04-12T15:11:06Z
001005756 915__ $$0StatID:(DE-HGF)0030$$2StatID$$aPeer Review$$bDOAJ : Anonymous peer review$$d2023-04-12T15:11:06Z
001005756 915__ $$0StatID:(DE-HGF)0600$$2StatID$$aDBCoverage$$bEbsco Academic Search$$d2023-08-24
001005756 915__ $$0StatID:(DE-HGF)0030$$2StatID$$aPeer Review$$bASC$$d2023-08-24
001005756 915__ $$0StatID:(DE-HGF)0199$$2StatID$$aDBCoverage$$bClarivate Analytics Master Journal List$$d2023-08-24
001005756 915__ $$0StatID:(DE-HGF)1050$$2StatID$$aDBCoverage$$bBIOSIS Previews$$d2023-08-24
001005756 915__ $$0StatID:(DE-HGF)0150$$2StatID$$aDBCoverage$$bWeb of Science Core Collection$$d2023-08-24
001005756 915__ $$0StatID:(DE-HGF)1040$$2StatID$$aDBCoverage$$bZoological Record$$d2023-08-24
001005756 915__ $$0StatID:(DE-HGF)1150$$2StatID$$aDBCoverage$$bCurrent Contents - Physical, Chemical and Earth Sciences$$d2023-08-24
001005756 915__ $$0StatID:(DE-HGF)9900$$2StatID$$aIF < 5$$d2023-08-24
001005756 920__ $$lno
001005756 9201_ $$0I:(DE-Juel1)JSC-20090406$$kJSC$$lJülich Supercomputing Center$$x0
001005756 980__ $$ajournal
001005756 980__ $$aVDB
001005756 980__ $$aUNRESTRICTED
001005756 980__ $$aI:(DE-Juel1)JSC-20090406
001005756 9801_ $$aFullTexts