Matches in SemOpenAlex for { <https://semopenalex.org/work/W4320175545> ?p ?o ?g. }
Showing items 1 to 65 of
65
with 100 items per page.
- W4320175545 abstract "The lack of generalizability of deep learning approaches for the automated diagnosis of pathologies in Wireless Capsule Endoscopy (WCE) has prevented any significant advantages from trickling down to real clinical practices. As a result, disease management using WCE continues to depend on exhaustive manual investigations by medical experts. This explains its limited use despite several advantages. Prior works have considered using higher quality and quantity of labels as a way of tackling the lack of generalization, however this is hardly scalable considering pathology diversity not to mention that labeling large datasets encumbers the medical staff additionally. We propose using freely available domain knowledge as priors to learn more robust and generalizable representations. We experimentally show that domain priors can benefit representations by acting in proxy of labels, thereby significantly reducing the labeling requirement while still enabling fully unsupervised yet pathology-aware learning. We use the contrastive objective along with prior-guided views during pretraining, where the view choices inspire sensitivity to pathological information. Extensive experiments on three datasets show that our method performs better than (or closes gap with) the state-of-the-art in the domain, establishing a new benchmark in pathology classification and cross-dataset generalization, as well as scaling to unseen pathology categories." @default.
- W4320175545 created "2023-02-13" @default.
- W4320175545 creator A5014677722 @default.
- W4320175545 creator A5063755742 @default.
- W4320175545 creator A5088359880 @default.
- W4320175545 date "2022-06-10" @default.
- W4320175545 modified "2023-10-18" @default.
- W4320175545 title "From Labels to Priors in Capsule Endoscopy: A Prior Guided Approach for Improving Generalization with Few Labels" @default.
- W4320175545 doi "https://doi.org/10.48550/arxiv.2206.05288" @default.
- W4320175545 hasPublicationYear "2022" @default.
- W4320175545 type Work @default.
- W4320175545 citedByCount "0" @default.
- W4320175545 crossrefType "posted-content" @default.
- W4320175545 hasAuthorship W4320175545A5014677722 @default.
- W4320175545 hasAuthorship W4320175545A5063755742 @default.
- W4320175545 hasAuthorship W4320175545A5088359880 @default.
- W4320175545 hasBestOaLocation W43201755451 @default.
- W4320175545 hasConcept C105795698 @default.
- W4320175545 hasConcept C107673813 @default.
- W4320175545 hasConcept C119857082 @default.
- W4320175545 hasConcept C13280743 @default.
- W4320175545 hasConcept C134306372 @default.
- W4320175545 hasConcept C154945302 @default.
- W4320175545 hasConcept C177148314 @default.
- W4320175545 hasConcept C177769412 @default.
- W4320175545 hasConcept C185798385 @default.
- W4320175545 hasConcept C205649164 @default.
- W4320175545 hasConcept C27158222 @default.
- W4320175545 hasConcept C33923547 @default.
- W4320175545 hasConcept C36503486 @default.
- W4320175545 hasConcept C41008148 @default.
- W4320175545 hasConcept C48044578 @default.
- W4320175545 hasConcept C77088390 @default.
- W4320175545 hasConceptScore W4320175545C105795698 @default.
- W4320175545 hasConceptScore W4320175545C107673813 @default.
- W4320175545 hasConceptScore W4320175545C119857082 @default.
- W4320175545 hasConceptScore W4320175545C13280743 @default.
- W4320175545 hasConceptScore W4320175545C134306372 @default.
- W4320175545 hasConceptScore W4320175545C154945302 @default.
- W4320175545 hasConceptScore W4320175545C177148314 @default.
- W4320175545 hasConceptScore W4320175545C177769412 @default.
- W4320175545 hasConceptScore W4320175545C185798385 @default.
- W4320175545 hasConceptScore W4320175545C205649164 @default.
- W4320175545 hasConceptScore W4320175545C27158222 @default.
- W4320175545 hasConceptScore W4320175545C33923547 @default.
- W4320175545 hasConceptScore W4320175545C36503486 @default.
- W4320175545 hasConceptScore W4320175545C41008148 @default.
- W4320175545 hasConceptScore W4320175545C48044578 @default.
- W4320175545 hasConceptScore W4320175545C77088390 @default.
- W4320175545 hasLocation W43201755451 @default.
- W4320175545 hasOpenAccess W4320175545 @default.
- W4320175545 hasPrimaryLocation W43201755451 @default.
- W4320175545 hasRelatedWork W112744582 @default.
- W4320175545 hasRelatedWork W1992807924 @default.
- W4320175545 hasRelatedWork W2561082235 @default.
- W4320175545 hasRelatedWork W2789601449 @default.
- W4320175545 hasRelatedWork W2989932438 @default.
- W4320175545 hasRelatedWork W3014160944 @default.
- W4320175545 hasRelatedWork W3088526462 @default.
- W4320175545 hasRelatedWork W3175415172 @default.
- W4320175545 hasRelatedWork W4221150964 @default.
- W4320175545 hasRelatedWork W4312365371 @default.
- W4320175545 isParatext "false" @default.
- W4320175545 isRetracted "false" @default.
- W4320175545 workType "article" @default.