Matches in SemOpenAlex for { <https://semopenalex.org/work/W4387211792> ?p ?o ?g. }
Showing items 1 to 86 of
86
with 100 items per page.
- W4387211792 endingPage "367" @default.
- W4387211792 startingPage "358" @default.
- W4387211792 abstract "The automatic classification of 3D medical data is memory-intensive. Also, variations in the number of slices between samples is common. Naïve solutions such as subsampling can solve these problems, but at the cost of potentially eliminating relevant diagnosis information. Transformers have shown promising performance for sequential data analysis. However, their application for long sequences is data, computationally, and memory demanding. In this paper, we propose an end-to-end Transformer-based framework that allows to classify volumetric data of variable length in an efficient fashion. Particularly, by randomizing the input volume-wise resolution(#slices) during training, we enhance the capacity of the learnable positional embedding assigned to each volume slice. Consequently, the accumulated positional information in each positional embedding can be generalized to the neighbouring slices, even for high-resolution volumes at the test time. By doing so, the model will be more robust to variable volume length and amenable to different computational budgets. We evaluated the proposed approach in retinal OCT volume classification and achieved 21.96 $$%$$ average improvement in balanced accuracy on a 9-class diagnostic task, compared to state-of-the-art video transformers. Our findings show that varying the volume-wise resolution of the input during training results in more informative volume representation as compared to training with fixed number of slices per volume." @default.
- W4387211792 created "2023-10-01" @default.
- W4387211792 creator A5027079272 @default.
- W4387211792 creator A5038536873 @default.
- W4387211792 creator A5039727719 @default.
- W4387211792 creator A5053563726 @default.
- W4387211792 creator A5079021497 @default.
- W4387211792 date "2023-01-01" @default.
- W4387211792 modified "2023-10-01" @default.
- W4387211792 title "Transformer-Based End-to-End Classification of Variable-Length Volumetric Data" @default.
- W4387211792 cites W2109453625 @default.
- W4387211792 cites W2594690981 @default.
- W4387211792 cites W2769254490 @default.
- W4387211792 cites W2773475231 @default.
- W4387211792 cites W2886281300 @default.
- W4387211792 cites W2897158445 @default.
- W4387211792 cites W2953040471 @default.
- W4387211792 cites W2962710043 @default.
- W4387211792 cites W2990987289 @default.
- W4387211792 cites W2998508940 @default.
- W4387211792 cites W3033178972 @default.
- W4387211792 cites W3035267940 @default.
- W4387211792 cites W3036901136 @default.
- W4387211792 cites W3083622693 @default.
- W4387211792 cites W3087413301 @default.
- W4387211792 cites W4214612132 @default.
- W4387211792 cites W4295025017 @default.
- W4387211792 cites W4295934551 @default.
- W4387211792 cites W4295940432 @default.
- W4387211792 cites W4386075535 @default.
- W4387211792 doi "https://doi.org/10.1007/978-3-031-43987-2_35" @default.
- W4387211792 hasPublicationYear "2023" @default.
- W4387211792 type Work @default.
- W4387211792 citedByCount "0" @default.
- W4387211792 crossrefType "book-chapter" @default.
- W4387211792 hasAuthorship W4387211792A5027079272 @default.
- W4387211792 hasAuthorship W4387211792A5038536873 @default.
- W4387211792 hasAuthorship W4387211792A5039727719 @default.
- W4387211792 hasAuthorship W4387211792A5053563726 @default.
- W4387211792 hasAuthorship W4387211792A5079021497 @default.
- W4387211792 hasConcept C11413529 @default.
- W4387211792 hasConcept C121332964 @default.
- W4387211792 hasConcept C134306372 @default.
- W4387211792 hasConcept C153180895 @default.
- W4387211792 hasConcept C154945302 @default.
- W4387211792 hasConcept C165801399 @default.
- W4387211792 hasConcept C182365436 @default.
- W4387211792 hasConcept C20556612 @default.
- W4387211792 hasConcept C33923547 @default.
- W4387211792 hasConcept C41008148 @default.
- W4387211792 hasConcept C41608201 @default.
- W4387211792 hasConcept C51632099 @default.
- W4387211792 hasConcept C62520636 @default.
- W4387211792 hasConcept C66322947 @default.
- W4387211792 hasConceptScore W4387211792C11413529 @default.
- W4387211792 hasConceptScore W4387211792C121332964 @default.
- W4387211792 hasConceptScore W4387211792C134306372 @default.
- W4387211792 hasConceptScore W4387211792C153180895 @default.
- W4387211792 hasConceptScore W4387211792C154945302 @default.
- W4387211792 hasConceptScore W4387211792C165801399 @default.
- W4387211792 hasConceptScore W4387211792C182365436 @default.
- W4387211792 hasConceptScore W4387211792C20556612 @default.
- W4387211792 hasConceptScore W4387211792C33923547 @default.
- W4387211792 hasConceptScore W4387211792C41008148 @default.
- W4387211792 hasConceptScore W4387211792C41608201 @default.
- W4387211792 hasConceptScore W4387211792C51632099 @default.
- W4387211792 hasConceptScore W4387211792C62520636 @default.
- W4387211792 hasConceptScore W4387211792C66322947 @default.
- W4387211792 hasLocation W43872117921 @default.
- W4387211792 hasOpenAccess W4387211792 @default.
- W4387211792 hasPrimaryLocation W43872117921 @default.
- W4387211792 hasRelatedWork W2033914206 @default.
- W4387211792 hasRelatedWork W2042327336 @default.
- W4387211792 hasRelatedWork W2046077695 @default.
- W4387211792 hasRelatedWork W2055709700 @default.
- W4387211792 hasRelatedWork W2146076056 @default.
- W4387211792 hasRelatedWork W2163831990 @default.
- W4387211792 hasRelatedWork W2378160586 @default.
- W4387211792 hasRelatedWork W2996038082 @default.
- W4387211792 hasRelatedWork W3003836766 @default.
- W4387211792 hasRelatedWork W3047965787 @default.
- W4387211792 isParatext "false" @default.
- W4387211792 isRetracted "false" @default.
- W4387211792 workType "book-chapter" @default.