FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3129733985> ?p ?o ?g. }

• W3129733985 abstract "ABSTRACT In response to the COVID-19 global pandemic, recent research has proposed creating deep learning based models that use chest radiographs (CXRs) in a variety of clinical tasks to help manage the crisis. However, the size of existing datasets of CXRs from COVID-19+ patients are relatively small, and researchers often pool CXR data from multiple sources, for example, using different x-ray machines in various patient populations under different clinical scenarios. Deep learning models trained on such datasets have been shown to overfit to erroneous features instead of learning pulmonary characteristics – a phenomenon known as shortcut learning. We propose adding feature disentanglement to the training process, forcing the models to identify pulmonary features from the images while penalizing them for learning features that can discriminate between the original datasets that the images come from. We find that models trained in this way indeed have better generalization performance on unseen data; in the best case we found that it improved AUC by 0.13 on held out data. We further find that this outperforms masking out non-lung parts of the CXRs and performing histogram equalization, both of which are recently proposed methods for removing biases in CXR datasets." @default.
• W3129733985 created "2021-03-01" @default.
• W3129733985 creator A5016793616 @default.
• W3129733985 creator A5018606314 @default.
• W3129733985 creator A5021432201 @default.
• W3129733985 creator A5022013110 @default.
• W3129733985 creator A5029283104 @default.
• W3129733985 creator A5032803955 @default.
• W3129733985 creator A5039535916 @default.
• W3129733985 creator A5044405868 @default.
• W3129733985 creator A5051295899 @default.
• W3129733985 creator A5064975442 @default.
• W3129733985 creator A5066063583 @default.
• W3129733985 creator A5088583724 @default.
• W3129733985 date "2021-02-13" @default.
• W3129733985 modified "2023-10-18" @default.
• W3129733985 title "Deep learning models for COVID-19 chest x-ray classification: Preventing shortcut learning using feature disentanglement" @default.
• W3129733985 cites W2752782242 @default.
• W3129733985 cites W3010702679 @default.
• W3129733985 cites W3013601031 @default.
• W3129733985 cites W3014129273 @default.
• W3129733985 cites W3014524604 @default.
• W3129733985 cites W3015984951 @default.
• W3129733985 cites W3017117984 @default.
• W3129733985 cites W3017855299 @default.
• W3129733985 cites W3020653337 @default.
• W3129733985 cites W3021001507 @default.
• W3129733985 cites W3032747197 @default.
• W3129733985 cites W3041809298 @default.
• W3129733985 cites W3045460727 @default.
• W3129733985 cites W3084392726 @default.
• W3129733985 cites W3085109610 @default.
• W3129733985 cites W3105081694 @default.
• W3129733985 doi "https://doi.org/10.1101/2021.02.11.20196766" @default.
• W3129733985 hasPubMedCentralId "https://www.ncbi.nlm.nih.gov/pmc/articles/7885941" @default.
• W3129733985 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/33594382" @default.
• W3129733985 hasPublicationYear "2021" @default.
• W3129733985 type Work @default.
• W3129733985 sameAs 3129733985 @default.
• W3129733985 citedByCount "5" @default.
• W3129733985 countsByYear W31297339852021 @default.
• W3129733985 countsByYear W31297339852022 @default.
• W3129733985 countsByYear W31297339852023 @default.
• W3129733985 crossrefType "posted-content" @default.
• W3129733985 hasAuthorship W3129733985A5016793616 @default.
• W3129733985 hasAuthorship W3129733985A5018606314 @default.
• W3129733985 hasAuthorship W3129733985A5021432201 @default.
• W3129733985 hasAuthorship W3129733985A5022013110 @default.
• W3129733985 hasAuthorship W3129733985A5029283104 @default.
• W3129733985 hasAuthorship W3129733985A5032803955 @default.
• W3129733985 hasAuthorship W3129733985A5039535916 @default.
• W3129733985 hasAuthorship W3129733985A5044405868 @default.
• W3129733985 hasAuthorship W3129733985A5051295899 @default.
• W3129733985 hasAuthorship W3129733985A5064975442 @default.
• W3129733985 hasAuthorship W3129733985A5066063583 @default.
• W3129733985 hasAuthorship W3129733985A5088583724 @default.
• W3129733985 hasBestOaLocation W31297339851 @default.
• W3129733985 hasConcept C108583219 @default.
• W3129733985 hasConcept C119857082 @default.
• W3129733985 hasConcept C134306372 @default.
• W3129733985 hasConcept C138885662 @default.
• W3129733985 hasConcept C142362112 @default.
• W3129733985 hasConcept C142724271 @default.
• W3129733985 hasConcept C150899416 @default.
• W3129733985 hasConcept C153180895 @default.
• W3129733985 hasConcept C153349607 @default.
• W3129733985 hasConcept C154945302 @default.
• W3129733985 hasConcept C177148314 @default.
• W3129733985 hasConcept C197115733 @default.
• W3129733985 hasConcept C22019652 @default.
• W3129733985 hasConcept C2776401178 @default.
• W3129733985 hasConcept C2777402240 @default.
• W3129733985 hasConcept C2779134260 @default.
• W3129733985 hasConcept C3008058167 @default.
• W3129733985 hasConcept C33923547 @default.
• W3129733985 hasConcept C41008148 @default.
• W3129733985 hasConcept C41895202 @default.
• W3129733985 hasConcept C50644808 @default.
• W3129733985 hasConcept C524204448 @default.
• W3129733985 hasConcept C59404180 @default.
• W3129733985 hasConcept C71924100 @default.
• W3129733985 hasConceptScore W3129733985C108583219 @default.
• W3129733985 hasConceptScore W3129733985C119857082 @default.
• W3129733985 hasConceptScore W3129733985C134306372 @default.
• W3129733985 hasConceptScore W3129733985C138885662 @default.
• W3129733985 hasConceptScore W3129733985C142362112 @default.
• W3129733985 hasConceptScore W3129733985C142724271 @default.
• W3129733985 hasConceptScore W3129733985C150899416 @default.
• W3129733985 hasConceptScore W3129733985C153180895 @default.
• W3129733985 hasConceptScore W3129733985C153349607 @default.
• W3129733985 hasConceptScore W3129733985C154945302 @default.
• W3129733985 hasConceptScore W3129733985C177148314 @default.
• W3129733985 hasConceptScore W3129733985C197115733 @default.
• W3129733985 hasConceptScore W3129733985C22019652 @default.
• W3129733985 hasConceptScore W3129733985C2776401178 @default.
• W3129733985 hasConceptScore W3129733985C2777402240 @default.
• W3129733985 hasConceptScore W3129733985C2779134260 @default.
• W3129733985 hasConceptScore W3129733985C3008058167 @default.
• W3129733985 hasConceptScore W3129733985C33923547 @default.
• W3129733985 hasConceptScore W3129733985C41008148 @default.

• next