Matches in SemOpenAlex for { <https://semopenalex.org/work/W4301183166> ?p ?o ?g. }
Showing items 1 to 54 of
54
with 100 items per page.
- W4301183166 abstract "Neural network training and validation rely on the availability of large high-quality datasets. However, in many cases only incomplete datasets are available, particularly in health care applications, where each patient typically undergoes different clinical procedures or can drop out of a study. Here, we introduce GapNet, an alternative deep-learning training approach that can use highly incomplete datasets without overfitting or introducing artefacts. Using two highly incomplete real-world medical datasets, we show that GapNet improves the identification of patients with underlying Alzheimer's disease pathology and of patients at risk of hospitalization due to Covid-19. Compared to commonly used imputation methods, this improvement suggests that GapNet can become a general tool to handle incomplete medical datasets." @default.
- W4301183166 created "2022-10-04" @default.
- W4301183166 creator A5035357435 @default.
- W4301183166 date "2022-10-04" @default.
- W4301183166 modified "2023-09-28" @default.
- W4301183166 title "Neural network training with highly incomplete medical datasets (Conference Presentation)" @default.
- W4301183166 doi "https://doi.org/10.1117/12.2655654" @default.
- W4301183166 hasPublicationYear "2022" @default.
- W4301183166 type Work @default.
- W4301183166 citedByCount "0" @default.
- W4301183166 crossrefType "proceedings-article" @default.
- W4301183166 hasAuthorship W4301183166A5035357435 @default.
- W4301183166 hasConcept C108583219 @default.
- W4301183166 hasConcept C116834253 @default.
- W4301183166 hasConcept C119857082 @default.
- W4301183166 hasConcept C124101348 @default.
- W4301183166 hasConcept C154945302 @default.
- W4301183166 hasConcept C22019652 @default.
- W4301183166 hasConcept C2984842247 @default.
- W4301183166 hasConcept C41008148 @default.
- W4301183166 hasConcept C50644808 @default.
- W4301183166 hasConcept C58041806 @default.
- W4301183166 hasConcept C59822182 @default.
- W4301183166 hasConcept C86803240 @default.
- W4301183166 hasConcept C9357733 @default.
- W4301183166 hasConceptScore W4301183166C108583219 @default.
- W4301183166 hasConceptScore W4301183166C116834253 @default.
- W4301183166 hasConceptScore W4301183166C119857082 @default.
- W4301183166 hasConceptScore W4301183166C124101348 @default.
- W4301183166 hasConceptScore W4301183166C154945302 @default.
- W4301183166 hasConceptScore W4301183166C22019652 @default.
- W4301183166 hasConceptScore W4301183166C2984842247 @default.
- W4301183166 hasConceptScore W4301183166C41008148 @default.
- W4301183166 hasConceptScore W4301183166C50644808 @default.
- W4301183166 hasConceptScore W4301183166C58041806 @default.
- W4301183166 hasConceptScore W4301183166C59822182 @default.
- W4301183166 hasConceptScore W4301183166C86803240 @default.
- W4301183166 hasConceptScore W4301183166C9357733 @default.
- W4301183166 hasLocation W43011831661 @default.
- W4301183166 hasOpenAccess W4301183166 @default.
- W4301183166 hasPrimaryLocation W43011831661 @default.
- W4301183166 hasRelatedWork W1530536511 @default.
- W4301183166 hasRelatedWork W3099765033 @default.
- W4301183166 hasRelatedWork W3186840088 @default.
- W4301183166 hasRelatedWork W3186919929 @default.
- W4301183166 hasRelatedWork W4285802257 @default.
- W4301183166 hasRelatedWork W4287064118 @default.
- W4301183166 hasRelatedWork W4308233823 @default.
- W4301183166 hasRelatedWork W4360619137 @default.
- W4301183166 hasRelatedWork W4361732492 @default.
- W4301183166 hasRelatedWork W4362499066 @default.
- W4301183166 isParatext "false" @default.
- W4301183166 isRetracted "false" @default.
- W4301183166 workType "article" @default.