Matches in SemOpenAlex for { <https://semopenalex.org/work/W4367046802> ?p ?o ?g. }
- W4367046802 abstract "Deep learning techniques are capable of capturing complex input-output relationships, and have been widely applied to the diagnosis prediction task based on web-based patient electronic health records (EHR) data. To improve the prediction and interpretability of pure data-driven deep learning with only a limited amount of labeled data, a pervasive trend is to assist the model training with knowledge priors from online medical knowledge graphs. However, they marginally investigated the label imbalance and the task-irrelevant noise in the external knowledge graph. The imbalanced label distribution would bias the learning and knowledge extraction towards the majority categories. The task-irrelevant noise introduces extra uncertainty to the model performance. To this end, aiming at by-passing the bias-variance trade-off dilemma, we introduce a new sequential learning framework, dubbed SeqCare, for diagnosis prediction with online medical knowledge graphs. Concretely, in the first step, SeqCare learns a bias-reduced space through a self-supervised graph contrastive learning task. Secondly, SeqCare reduces the learning uncertainty by refining the supervision signal and the graph structure of the knowledge graph simultaneously. Lastly, SeqCare trains the model in the bias-variance reduced space with a self-distillation to further filter out irrelevant information in the data. Experimental evaluations on two real-world datasets show that SeqCare outperforms state-of-the-art approaches. Case studies exemplify the interpretability of SeqCare. Moreover, the medical findings discovered by SeqCare are consistent with experts and medical literature." @default.
- W4367046802 created "2023-04-27" @default.
- W4367046802 creator A5007593210 @default.
- W4367046802 creator A5007959870 @default.
- W4367046802 creator A5009005391 @default.
- W4367046802 creator A5009946754 @default.
- W4367046802 creator A5012305964 @default.
- W4367046802 creator A5026097557 @default.
- W4367046802 creator A5035448666 @default.
- W4367046802 creator A5055336632 @default.
- W4367046802 creator A5064595835 @default.
- W4367046802 date "2023-04-30" @default.
- W4367046802 modified "2023-10-16" @default.
- W4367046802 title "SeqCare: Sequential Training with External Medical Knowledge Graph for Diagnosis Prediction in Healthcare Data" @default.
- W4367046802 cites W1969116741 @default.
- W4367046802 cites W1976526581 @default.
- W4367046802 cites W2023450550 @default.
- W4367046802 cites W2106897586 @default.
- W4367046802 cites W2132791018 @default.
- W4367046802 cites W2142177680 @default.
- W4367046802 cites W2148143831 @default.
- W4367046802 cites W2179187436 @default.
- W4367046802 cites W2557074642 @default.
- W4367046802 cites W2604836086 @default.
- W4367046802 cites W2742491462 @default.
- W4367046802 cites W2896538705 @default.
- W4367046802 cites W2913198899 @default.
- W4367046802 cites W2963351448 @default.
- W4367046802 cites W2963532813 @default.
- W4367046802 cites W2963691377 @default.
- W4367046802 cites W2997653844 @default.
- W4367046802 cites W2998301895 @default.
- W4367046802 cites W2998409174 @default.
- W4367046802 cites W3003265726 @default.
- W4367046802 cites W3010336026 @default.
- W4367046802 cites W3012847979 @default.
- W4367046802 cites W3022196370 @default.
- W4367046802 cites W3035286001 @default.
- W4367046802 cites W3080098168 @default.
- W4367046802 cites W3093599560 @default.
- W4367046802 cites W3095707208 @default.
- W4367046802 cites W3099136959 @default.
- W4367046802 cites W3101227480 @default.
- W4367046802 cites W3126752450 @default.
- W4367046802 cites W3153405813 @default.
- W4367046802 cites W3156039010 @default.
- W4367046802 cites W3156877449 @default.
- W4367046802 cites W3166596953 @default.
- W4367046802 cites W3177373898 @default.
- W4367046802 cites W3204988419 @default.
- W4367046802 cites W4290944617 @default.
- W4367046802 doi "https://doi.org/10.1145/3543507.3583543" @default.
- W4367046802 hasPublicationYear "2023" @default.
- W4367046802 type Work @default.
- W4367046802 citedByCount "0" @default.
- W4367046802 crossrefType "proceedings-article" @default.
- W4367046802 hasAuthorship W4367046802A5007593210 @default.
- W4367046802 hasAuthorship W4367046802A5007959870 @default.
- W4367046802 hasAuthorship W4367046802A5009005391 @default.
- W4367046802 hasAuthorship W4367046802A5009946754 @default.
- W4367046802 hasAuthorship W4367046802A5012305964 @default.
- W4367046802 hasAuthorship W4367046802A5026097557 @default.
- W4367046802 hasAuthorship W4367046802A5035448666 @default.
- W4367046802 hasAuthorship W4367046802A5055336632 @default.
- W4367046802 hasAuthorship W4367046802A5064595835 @default.
- W4367046802 hasConcept C108583219 @default.
- W4367046802 hasConcept C11413529 @default.
- W4367046802 hasConcept C119857082 @default.
- W4367046802 hasConcept C124101348 @default.
- W4367046802 hasConcept C132525143 @default.
- W4367046802 hasConcept C154945302 @default.
- W4367046802 hasConcept C2781067378 @default.
- W4367046802 hasConcept C41008148 @default.
- W4367046802 hasConcept C74197172 @default.
- W4367046802 hasConcept C80444323 @default.
- W4367046802 hasConceptScore W4367046802C108583219 @default.
- W4367046802 hasConceptScore W4367046802C11413529 @default.
- W4367046802 hasConceptScore W4367046802C119857082 @default.
- W4367046802 hasConceptScore W4367046802C124101348 @default.
- W4367046802 hasConceptScore W4367046802C132525143 @default.
- W4367046802 hasConceptScore W4367046802C154945302 @default.
- W4367046802 hasConceptScore W4367046802C2781067378 @default.
- W4367046802 hasConceptScore W4367046802C41008148 @default.
- W4367046802 hasConceptScore W4367046802C74197172 @default.
- W4367046802 hasConceptScore W4367046802C80444323 @default.
- W4367046802 hasLocation W43670468021 @default.
- W4367046802 hasOpenAccess W4367046802 @default.
- W4367046802 hasPrimaryLocation W43670468021 @default.
- W4367046802 hasRelatedWork W2605281151 @default.
- W4367046802 hasRelatedWork W3006943036 @default.
- W4367046802 hasRelatedWork W3191046242 @default.
- W4367046802 hasRelatedWork W3208423683 @default.
- W4367046802 hasRelatedWork W4206493799 @default.
- W4367046802 hasRelatedWork W4206534706 @default.
- W4367046802 hasRelatedWork W4213225422 @default.
- W4367046802 hasRelatedWork W4213432687 @default.
- W4367046802 hasRelatedWork W4294031299 @default.
- W4367046802 hasRelatedWork W4299487748 @default.
- W4367046802 isParatext "false" @default.
- W4367046802 isRetracted "false" @default.