Matches in SemOpenAlex for { <https://semopenalex.org/work/W2975098105> ?p ?o ?g. }
Showing items 1 to 88 of
88
with 100 items per page.
- W2975098105 abstract "Missing data is a common problem in machine learning and in retrospective imaging research it is often encountered in the form of missing imaging modalities. We propose to take into account missing modalities in the design and training of neural networks, to ensure that they are capable of providing the best possible prediction even when multiple images are not available. The proposed network combines three modifications to the standard 3D UNet architecture: a training scheme with dropout of modalities, a multi-pathway architecture with fusion layer in the final stage, and the separate pre-training of these pathways. These modifications are evaluated incrementally in terms of performance on full and missing data, using the BraTS multi-modal segmentation challenge. The final model shows significant improvement with respect to the state of the art on missing data and requires less memory during training." @default.
- W2975098105 created "2019-10-03" @default.
- W2975098105 creator A5027857046 @default.
- W2975098105 creator A5042597698 @default.
- W2975098105 creator A5052901925 @default.
- W2975098105 date "2019-09-25" @default.
- W2975098105 modified "2023-09-27" @default.
- W2975098105 title "Multi-modal segmentation with missing MR sequences using pre-trained fusion networks" @default.
- W2975098105 cites W1641498739 @default.
- W2975098105 cites W1977185509 @default.
- W2975098105 cites W2187089797 @default.
- W2975098105 cites W2397287600 @default.
- W2975098105 cites W2464708700 @default.
- W2975098105 cites W2502805798 @default.
- W2975098105 cites W2731173046 @default.
- W2975098105 cites W2751069891 @default.
- W2975098105 cites W2964121744 @default.
- W2975098105 hasPublicationYear "2019" @default.
- W2975098105 type Work @default.
- W2975098105 sameAs 2975098105 @default.
- W2975098105 citedByCount "0" @default.
- W2975098105 crossrefType "posted-content" @default.
- W2975098105 hasAuthorship W2975098105A5027857046 @default.
- W2975098105 hasAuthorship W2975098105A5042597698 @default.
- W2975098105 hasAuthorship W2975098105A5052901925 @default.
- W2975098105 hasConcept C119857082 @default.
- W2975098105 hasConcept C134306372 @default.
- W2975098105 hasConcept C144024400 @default.
- W2975098105 hasConcept C153180895 @default.
- W2975098105 hasConcept C154945302 @default.
- W2975098105 hasConcept C185592680 @default.
- W2975098105 hasConcept C188027245 @default.
- W2975098105 hasConcept C2776145597 @default.
- W2975098105 hasConcept C2779903281 @default.
- W2975098105 hasConcept C2780226545 @default.
- W2975098105 hasConcept C33923547 @default.
- W2975098105 hasConcept C36289849 @default.
- W2975098105 hasConcept C41008148 @default.
- W2975098105 hasConcept C50644808 @default.
- W2975098105 hasConcept C71139939 @default.
- W2975098105 hasConcept C77618280 @default.
- W2975098105 hasConcept C89600930 @default.
- W2975098105 hasConcept C9357733 @default.
- W2975098105 hasConceptScore W2975098105C119857082 @default.
- W2975098105 hasConceptScore W2975098105C134306372 @default.
- W2975098105 hasConceptScore W2975098105C144024400 @default.
- W2975098105 hasConceptScore W2975098105C153180895 @default.
- W2975098105 hasConceptScore W2975098105C154945302 @default.
- W2975098105 hasConceptScore W2975098105C185592680 @default.
- W2975098105 hasConceptScore W2975098105C188027245 @default.
- W2975098105 hasConceptScore W2975098105C2776145597 @default.
- W2975098105 hasConceptScore W2975098105C2779903281 @default.
- W2975098105 hasConceptScore W2975098105C2780226545 @default.
- W2975098105 hasConceptScore W2975098105C33923547 @default.
- W2975098105 hasConceptScore W2975098105C36289849 @default.
- W2975098105 hasConceptScore W2975098105C41008148 @default.
- W2975098105 hasConceptScore W2975098105C50644808 @default.
- W2975098105 hasConceptScore W2975098105C71139939 @default.
- W2975098105 hasConceptScore W2975098105C77618280 @default.
- W2975098105 hasConceptScore W2975098105C89600930 @default.
- W2975098105 hasConceptScore W2975098105C9357733 @default.
- W2975098105 hasLocation W29750981051 @default.
- W2975098105 hasOpenAccess W2975098105 @default.
- W2975098105 hasPrimaryLocation W29750981051 @default.
- W2975098105 hasRelatedWork W1775792793 @default.
- W2975098105 hasRelatedWork W2741295496 @default.
- W2975098105 hasRelatedWork W2773119425 @default.
- W2975098105 hasRelatedWork W2882978096 @default.
- W2975098105 hasRelatedWork W2913643062 @default.
- W2975098105 hasRelatedWork W2941736865 @default.
- W2975098105 hasRelatedWork W2967057970 @default.
- W2975098105 hasRelatedWork W2979984877 @default.
- W2975098105 hasRelatedWork W2995907173 @default.
- W2975098105 hasRelatedWork W3016830346 @default.
- W2975098105 hasRelatedWork W3034162269 @default.
- W2975098105 hasRelatedWork W3089870586 @default.
- W2975098105 hasRelatedWork W3097616280 @default.
- W2975098105 hasRelatedWork W3100331755 @default.
- W2975098105 hasRelatedWork W3103982318 @default.
- W2975098105 hasRelatedWork W3150579659 @default.
- W2975098105 hasRelatedWork W3162399999 @default.
- W2975098105 hasRelatedWork W3173665739 @default.
- W2975098105 hasRelatedWork W3175770036 @default.
- W2975098105 hasRelatedWork W3189405384 @default.
- W2975098105 isParatext "false" @default.
- W2975098105 isRetracted "false" @default.
- W2975098105 magId "2975098105" @default.
- W2975098105 workType "article" @default.