Matches in SemOpenAlex for { <https://semopenalex.org/work/W4224284596> ?p ?o ?g. }
Showing items 1 to 92 of
92
with 100 items per page.
- W4224284596 abstract "ABSTRACT Background Temporal distribution shift negatively impacts the performance of clinical prediction models over time. Pretraining foundation models using self-supervised learning on electronic health records (EHR) may be effective in acquiring informative global patterns that can improve the robustness of task-specific models. Objective To evaluate the utility of EHR foundation models in improving the in-distribution (ID) and out-of-distribution (OOD) performance of clinical prediction models. Methods The cohort consisted of adult inpatients admitted between 2009-2021. Gated recurrent unit (GRU)- and transformer (TRANS)-based foundation models were pretrained on EHR of patients admitted between 2009-2012 and were subsequently used to construct patient representations (CLMBR). These representations were used to learn logistic regression models (CLMBR GRU and CLMBR TRANS ) to predict hospital mortality, long length of stay, 30-day readmission, and ICU admission. We compared CLMBR GRU and CLMBR TRANS with baseline logistic regression models learned on count-based representations (count-LR) and end-to-end (ETE) GRU and transformer models in ID (2009-2012) and OOD (2013-2021) year groups. Performance was measured using area-under-the-receiver-operating-characteristic curve, area- under-the-precision-recall curve, and absolute calibration error. Results Models trained on CLMBR generally showed better discrimination relative to count-LR in both ID and OOD year groups. In addition, they often matched or were better than their ETE counterparts. Finally, foundation models’ performance in the self-supervised learning task tracked closely with the ID and OOD performance of the downstream models. Conclusions These results suggest that pretraining foundation models on electronic health records is a useful approach for developing clinical prediction models that perform well in the presence of temporal distribution shift." @default.
- W4224284596 created "2022-04-26" @default.
- W4224284596 creator A5010316094 @default.
- W4224284596 creator A5026965117 @default.
- W4224284596 creator A5028277225 @default.
- W4224284596 creator A5034895647 @default.
- W4224284596 creator A5041175834 @default.
- W4224284596 creator A5070470494 @default.
- W4224284596 creator A5079049199 @default.
- W4224284596 creator A5086281849 @default.
- W4224284596 creator A5086900250 @default.
- W4224284596 date "2022-04-19" @default.
- W4224284596 modified "2023-10-14" @default.
- W4224284596 title "EHR Foundation Models Improve Robustness in the Presence of Temporal Distribution Shift" @default.
- W4224284596 cites W1914974178 @default.
- W4224284596 cites W2028138594 @default.
- W4224284596 cites W2079776874 @default.
- W4224284596 cites W2157365260 @default.
- W4224284596 cites W2159583324 @default.
- W4224284596 cites W2604834158 @default.
- W4224284596 cites W2799695199 @default.
- W4224284596 cites W2954707123 @default.
- W4224284596 cites W2971616138 @default.
- W4224284596 cites W3043023066 @default.
- W4224284596 cites W3112116031 @default.
- W4224284596 cites W3174786846 @default.
- W4224284596 cites W3177617320 @default.
- W4224284596 cites W3196529262 @default.
- W4224284596 cites W4221077687 @default.
- W4224284596 doi "https://doi.org/10.1101/2022.04.15.22273900" @default.
- W4224284596 hasPublicationYear "2022" @default.
- W4224284596 type Work @default.
- W4224284596 citedByCount "0" @default.
- W4224284596 crossrefType "posted-content" @default.
- W4224284596 hasAuthorship W4224284596A5010316094 @default.
- W4224284596 hasAuthorship W4224284596A5026965117 @default.
- W4224284596 hasAuthorship W4224284596A5028277225 @default.
- W4224284596 hasAuthorship W4224284596A5034895647 @default.
- W4224284596 hasAuthorship W4224284596A5041175834 @default.
- W4224284596 hasAuthorship W4224284596A5070470494 @default.
- W4224284596 hasAuthorship W4224284596A5079049199 @default.
- W4224284596 hasAuthorship W4224284596A5086281849 @default.
- W4224284596 hasAuthorship W4224284596A5086900250 @default.
- W4224284596 hasBestOaLocation W42242845961 @default.
- W4224284596 hasConcept C104317684 @default.
- W4224284596 hasConcept C105795698 @default.
- W4224284596 hasConcept C119599485 @default.
- W4224284596 hasConcept C119857082 @default.
- W4224284596 hasConcept C127413603 @default.
- W4224284596 hasConcept C151956035 @default.
- W4224284596 hasConcept C154945302 @default.
- W4224284596 hasConcept C165801399 @default.
- W4224284596 hasConcept C185592680 @default.
- W4224284596 hasConcept C33923547 @default.
- W4224284596 hasConcept C41008148 @default.
- W4224284596 hasConcept C55493867 @default.
- W4224284596 hasConcept C58471807 @default.
- W4224284596 hasConcept C63479239 @default.
- W4224284596 hasConcept C66322947 @default.
- W4224284596 hasConcept C83546350 @default.
- W4224284596 hasConceptScore W4224284596C104317684 @default.
- W4224284596 hasConceptScore W4224284596C105795698 @default.
- W4224284596 hasConceptScore W4224284596C119599485 @default.
- W4224284596 hasConceptScore W4224284596C119857082 @default.
- W4224284596 hasConceptScore W4224284596C127413603 @default.
- W4224284596 hasConceptScore W4224284596C151956035 @default.
- W4224284596 hasConceptScore W4224284596C154945302 @default.
- W4224284596 hasConceptScore W4224284596C165801399 @default.
- W4224284596 hasConceptScore W4224284596C185592680 @default.
- W4224284596 hasConceptScore W4224284596C33923547 @default.
- W4224284596 hasConceptScore W4224284596C41008148 @default.
- W4224284596 hasConceptScore W4224284596C55493867 @default.
- W4224284596 hasConceptScore W4224284596C58471807 @default.
- W4224284596 hasConceptScore W4224284596C63479239 @default.
- W4224284596 hasConceptScore W4224284596C66322947 @default.
- W4224284596 hasConceptScore W4224284596C83546350 @default.
- W4224284596 hasLocation W42242845961 @default.
- W4224284596 hasOpenAccess W4224284596 @default.
- W4224284596 hasPrimaryLocation W42242845961 @default.
- W4224284596 hasRelatedWork W2127072394 @default.
- W4224284596 hasRelatedWork W2146073304 @default.
- W4224284596 hasRelatedWork W2799952019 @default.
- W4224284596 hasRelatedWork W3038347577 @default.
- W4224284596 hasRelatedWork W3047552631 @default.
- W4224284596 hasRelatedWork W3099386970 @default.
- W4224284596 hasRelatedWork W3174196512 @default.
- W4224284596 hasRelatedWork W4224214395 @default.
- W4224284596 hasRelatedWork W4249068542 @default.
- W4224284596 hasRelatedWork W4367596031 @default.
- W4224284596 isParatext "false" @default.
- W4224284596 isRetracted "false" @default.
- W4224284596 workType "article" @default.