Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386875088> ?p ?o ?g. }
Showing items 1 to 69 of
69
with 100 items per page.
- W4386875088 abstract "Since annotating medical images for segmentation tasks commonly incurs expensive costs, it is highly desirable to design an annotation-efficient method to alleviate the annotation burden. Recently, contrastive learning has exhibited a great potential in learning robust representations to boost downstream tasks with limited labels. In medical imaging scenarios, ready-made meta labels (i.e., specific attribute information of medical images) inherently reveal semantic relationships among images, which have been used to define positive pairs in previous work. However, the multi-perspective semantics revealed by various meta labels are usually incompatible and can incur intractable semantic contradiction when combining different meta labels. In this paper, we tackle the issue of semantic contradiction in a gradient-guided manner using our proposed Gradient Mitigator method, which systematically unifies multi-perspective meta labels to enable a pre-trained model to attain a better high-level semantic recognition ability. Moreover, we emphasize that the fine-grained discrimination ability is vital for segmentation-oriented pre-training, and develop a novel method called Gradient Filter to dynamically screen pixel pairs with the most discriminating power based on the magnitude of gradients. Comprehensive experiments on four medical image segmentation datasets verify that our new method GCL: (1) learns informative image representations and considerably boosts segmentation performance with limited labels, and (2) shows promising generalizability on out-of-distribution datasets." @default.
- W4386875088 created "2023-09-20" @default.
- W4386875088 creator A5014958566 @default.
- W4386875088 creator A5023258354 @default.
- W4386875088 creator A5029687287 @default.
- W4386875088 creator A5052610760 @default.
- W4386875088 creator A5060901632 @default.
- W4386875088 creator A5065115366 @default.
- W4386875088 date "2023-09-16" @default.
- W4386875088 modified "2023-09-27" @default.
- W4386875088 title "GCL: Gradient-Guided Contrastive Learning for Medical Image Segmentation with Multi-Perspective Meta Labels" @default.
- W4386875088 doi "https://doi.org/10.48550/arxiv.2309.08888" @default.
- W4386875088 hasPublicationYear "2023" @default.
- W4386875088 type Work @default.
- W4386875088 citedByCount "0" @default.
- W4386875088 crossrefType "posted-content" @default.
- W4386875088 hasAuthorship W4386875088A5014958566 @default.
- W4386875088 hasAuthorship W4386875088A5023258354 @default.
- W4386875088 hasAuthorship W4386875088A5029687287 @default.
- W4386875088 hasAuthorship W4386875088A5052610760 @default.
- W4386875088 hasAuthorship W4386875088A5060901632 @default.
- W4386875088 hasAuthorship W4386875088A5065115366 @default.
- W4386875088 hasBestOaLocation W43868750881 @default.
- W4386875088 hasConcept C105795698 @default.
- W4386875088 hasConcept C119857082 @default.
- W4386875088 hasConcept C124504099 @default.
- W4386875088 hasConcept C12713177 @default.
- W4386875088 hasConcept C153180895 @default.
- W4386875088 hasConcept C154945302 @default.
- W4386875088 hasConcept C184337299 @default.
- W4386875088 hasConcept C199360897 @default.
- W4386875088 hasConcept C22019652 @default.
- W4386875088 hasConcept C27158222 @default.
- W4386875088 hasConcept C2776321320 @default.
- W4386875088 hasConcept C33923547 @default.
- W4386875088 hasConcept C41008148 @default.
- W4386875088 hasConcept C50644808 @default.
- W4386875088 hasConcept C89600930 @default.
- W4386875088 hasConceptScore W4386875088C105795698 @default.
- W4386875088 hasConceptScore W4386875088C119857082 @default.
- W4386875088 hasConceptScore W4386875088C124504099 @default.
- W4386875088 hasConceptScore W4386875088C12713177 @default.
- W4386875088 hasConceptScore W4386875088C153180895 @default.
- W4386875088 hasConceptScore W4386875088C154945302 @default.
- W4386875088 hasConceptScore W4386875088C184337299 @default.
- W4386875088 hasConceptScore W4386875088C199360897 @default.
- W4386875088 hasConceptScore W4386875088C22019652 @default.
- W4386875088 hasConceptScore W4386875088C27158222 @default.
- W4386875088 hasConceptScore W4386875088C2776321320 @default.
- W4386875088 hasConceptScore W4386875088C33923547 @default.
- W4386875088 hasConceptScore W4386875088C41008148 @default.
- W4386875088 hasConceptScore W4386875088C50644808 @default.
- W4386875088 hasConceptScore W4386875088C89600930 @default.
- W4386875088 hasLocation W43868750881 @default.
- W4386875088 hasOpenAccess W4386875088 @default.
- W4386875088 hasPrimaryLocation W43868750881 @default.
- W4386875088 hasRelatedWork W1996541855 @default.
- W4386875088 hasRelatedWork W2742991909 @default.
- W4386875088 hasRelatedWork W2767651786 @default.
- W4386875088 hasRelatedWork W2899744278 @default.
- W4386875088 hasRelatedWork W2940336242 @default.
- W4386875088 hasRelatedWork W2989932438 @default.
- W4386875088 hasRelatedWork W3047749858 @default.
- W4386875088 hasRelatedWork W3099765033 @default.
- W4386875088 hasRelatedWork W4210794429 @default.
- W4386875088 hasRelatedWork W4283732135 @default.
- W4386875088 isParatext "false" @default.
- W4386875088 isRetracted "false" @default.
- W4386875088 workType "article" @default.