Matches in SemOpenAlex for { <https://semopenalex.org/work/W2979488462> ?p ?o ?g. }
Showing items 1 to 94 of
94
with 100 items per page.
- W2979488462 endingPage "684" @default.
- W2979488462 startingPage "676" @default.
- W2979488462 abstract "Deep neural networks have achieved tremendous success in image recognition, classification and object detection. However, deep learning is often criticised for its lack of transparency and general inability to rationalise its predictions. The issue of poor model interpretability becomes critical in medical applications: a model that is not understood and trusted by physicians is unlikely to be used in daily clinical practice. In this work, we develop a novel multi-task deep learning framework for simultaneous histopathology image classification and retrieval, leveraging on the classic concept of k-nearest neighbours to improve model interpretability. For a test image, we retrieve the most similar images from our training databases. These retrieved nearest neighbours can be used to classify the test image with a confidence score, and provide a human-interpretable explanation of our classification. Our original framework can be built on top of any existing classification network (and therefore benefit from pretrained models), by (i) combining a triplet loss function with a novel triplet sampling strategy to compare distances between samples and (ii) adding a Cauchy hashing loss function to accelerate neighbour searching. We evaluate our method on colorectal cancer histology slides and show that the confidence estimates are strongly correlated with model performance. Nearest neighbours are intuitive and useful for expert evaluation. They give insights into understanding possible model failures, and can support clinical decision making by comparing archived images and patient records with the actual case." @default.
- W2979488462 created "2019-10-18" @default.
- W2979488462 creator A5004138391 @default.
- W2979488462 creator A5018744564 @default.
- W2979488462 creator A5046721596 @default.
- W2979488462 creator A5046896448 @default.
- W2979488462 creator A5070636207 @default.
- W2979488462 date "2019-01-01" @default.
- W2979488462 modified "2023-09-30" @default.
- W2979488462 title "Multi-task Learning of a Deep K-Nearest Neighbour Network for Histopathological Image Classification and Retrieval" @default.
- W2979488462 cites W22040386 @default.
- W2979488462 cites W2411707397 @default.
- W2979488462 cites W2581082771 @default.
- W2979488462 cites W2760946358 @default.
- W2979488462 cites W2798834175 @default.
- W2979488462 cites W2890430415 @default.
- W2979488462 cites W2914568698 @default.
- W2979488462 cites W2963775347 @default.
- W2979488462 cites W2964280870 @default.
- W2979488462 cites W3099206234 @default.
- W2979488462 doi "https://doi.org/10.1007/978-3-030-32239-7_75" @default.
- W2979488462 hasPublicationYear "2019" @default.
- W2979488462 type Work @default.
- W2979488462 sameAs 2979488462 @default.
- W2979488462 citedByCount "20" @default.
- W2979488462 countsByYear W29794884622018 @default.
- W2979488462 countsByYear W29794884622020 @default.
- W2979488462 countsByYear W29794884622021 @default.
- W2979488462 countsByYear W29794884622022 @default.
- W2979488462 countsByYear W29794884622023 @default.
- W2979488462 crossrefType "book-chapter" @default.
- W2979488462 hasAuthorship W2979488462A5004138391 @default.
- W2979488462 hasAuthorship W2979488462A5018744564 @default.
- W2979488462 hasAuthorship W2979488462A5046721596 @default.
- W2979488462 hasAuthorship W2979488462A5046896448 @default.
- W2979488462 hasAuthorship W2979488462A5070636207 @default.
- W2979488462 hasBestOaLocation W29794884622 @default.
- W2979488462 hasConcept C108583219 @default.
- W2979488462 hasConcept C113238511 @default.
- W2979488462 hasConcept C115961682 @default.
- W2979488462 hasConcept C119857082 @default.
- W2979488462 hasConcept C124101348 @default.
- W2979488462 hasConcept C153180895 @default.
- W2979488462 hasConcept C154945302 @default.
- W2979488462 hasConcept C162324750 @default.
- W2979488462 hasConcept C1667742 @default.
- W2979488462 hasConcept C187736073 @default.
- W2979488462 hasConcept C2780451532 @default.
- W2979488462 hasConcept C2781067378 @default.
- W2979488462 hasConcept C2984842247 @default.
- W2979488462 hasConcept C38652104 @default.
- W2979488462 hasConcept C41008148 @default.
- W2979488462 hasConcept C50644808 @default.
- W2979488462 hasConcept C75294576 @default.
- W2979488462 hasConcept C99138194 @default.
- W2979488462 hasConceptScore W2979488462C108583219 @default.
- W2979488462 hasConceptScore W2979488462C113238511 @default.
- W2979488462 hasConceptScore W2979488462C115961682 @default.
- W2979488462 hasConceptScore W2979488462C119857082 @default.
- W2979488462 hasConceptScore W2979488462C124101348 @default.
- W2979488462 hasConceptScore W2979488462C153180895 @default.
- W2979488462 hasConceptScore W2979488462C154945302 @default.
- W2979488462 hasConceptScore W2979488462C162324750 @default.
- W2979488462 hasConceptScore W2979488462C1667742 @default.
- W2979488462 hasConceptScore W2979488462C187736073 @default.
- W2979488462 hasConceptScore W2979488462C2780451532 @default.
- W2979488462 hasConceptScore W2979488462C2781067378 @default.
- W2979488462 hasConceptScore W2979488462C2984842247 @default.
- W2979488462 hasConceptScore W2979488462C38652104 @default.
- W2979488462 hasConceptScore W2979488462C41008148 @default.
- W2979488462 hasConceptScore W2979488462C50644808 @default.
- W2979488462 hasConceptScore W2979488462C75294576 @default.
- W2979488462 hasConceptScore W2979488462C99138194 @default.
- W2979488462 hasLocation W29794884621 @default.
- W2979488462 hasLocation W29794884622 @default.
- W2979488462 hasLocation W29794884623 @default.
- W2979488462 hasOpenAccess W2979488462 @default.
- W2979488462 hasPrimaryLocation W29794884621 @default.
- W2979488462 hasRelatedWork W2399159263 @default.
- W2979488462 hasRelatedWork W2911294201 @default.
- W2979488462 hasRelatedWork W3006943036 @default.
- W2979488462 hasRelatedWork W3129898729 @default.
- W2979488462 hasRelatedWork W3208423683 @default.
- W2979488462 hasRelatedWork W4206079434 @default.
- W2979488462 hasRelatedWork W4206534706 @default.
- W2979488462 hasRelatedWork W4229079080 @default.
- W2979488462 hasRelatedWork W4299487748 @default.
- W2979488462 hasRelatedWork W4385957992 @default.
- W2979488462 isParatext "false" @default.
- W2979488462 isRetracted "false" @default.
- W2979488462 magId "2979488462" @default.
- W2979488462 workType "book-chapter" @default.