FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4221130850> ?p ?o ?g. }

• W4221130850 endingPage "105427" @default.
• W4221130850 startingPage "105427" @default.
• W4221130850 abstract "Owing to the data distribution shifts generated by collecting images using various imaging protocols and device vendors, the generalization capability of deep models is crucial for medical image analysis when applied to test datasets in clinical environments. Domain generalization (DG) methods have shown promising generalization performance in the field of medical image segmentation. In contrast to conventional DG, which has strict requirements regarding the availability of multiple source domains, we consider a more challenging problem, that is, single-domain generalization (SDG), where only a single source is available during network training. In this scenario, the augmentation of the entire image to improve the model generalization ability may cause alteration of hue values, resulting in the wrong segmentation of tissues in color medical images. To resolve this problem, we first present a novel illumination-randomized SDG framework to improve the model generalization power for color medical image segmentation by synthesizing randomized illumination maps. Specifically, we devise unsupervised retinex-based image decomposition neural networks (ID-Nets) to decompose color medical images into reflectance and illumination maps. Illumination maps are augmented by performing illumination randomization to generate medical color images under diverse illumination conditions. Second, to measure the quality of retinex-based image decomposition, we devise a novel metric, the transport gradient consistency index, by modeling physical illumination. Extensive experiments are performed to evaluate our proposed framework on two retinal fundus image segmentation tasks: optic cup and disc segmentation. The experimental results demonstrate that our framework outperforms other SDG and image enhancement methods, surpassing the state-of-the-art SDG methods by up to 9.6% with respect to the Dice coefficient." @default.
• W4221130850 created "2022-04-03" @default.
• W4221130850 creator A5023793169 @default.
• W4221130850 creator A5044602576 @default.
• W4221130850 creator A5061420652 @default.
• W4221130850 date "2022-06-01" @default.
• W4221130850 modified "2023-10-18" @default.
• W4221130850 title "Robust color medical image segmentation on unseen domain by randomized illumination enhancement" @default.
• W4221130850 cites W2076205488 @default.
• W4221130850 cites W2103559027 @default.
• W4221130850 cites W2115403315 @default.
• W4221130850 cites W2412782625 @default.
• W4221130850 cites W2566079294 @default.
• W4221130850 cites W2566376500 @default.
• W4221130850 cites W2589644515 @default.
• W4221130850 cites W2608353599 @default.
• W4221130850 cites W2623042903 @default.
• W4221130850 cites W2794103425 @default.
• W4221130850 cites W2899279931 @default.
• W4221130850 cites W2979448322 @default.
• W4221130850 cites W2979640220 @default.
• W4221130850 cites W3005498024 @default.
• W4221130850 cites W3006040295 @default.
• W4221130850 cites W3006326483 @default.
• W4221130850 cites W3048402654 @default.
• W4221130850 cites W3103010481 @default.
• W4221130850 cites W3120540810 @default.
• W4221130850 cites W3122431193 @default.
• W4221130850 cites W3136603779 @default.
• W4221130850 cites W3140251367 @default.
• W4221130850 cites W3142788400 @default.
• W4221130850 doi "https://doi.org/10.1016/j.compbiomed.2022.105427" @default.
• W4221130850 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/35585731" @default.
• W4221130850 hasPublicationYear "2022" @default.
• W4221130850 type Work @default.
• W4221130850 citedByCount "6" @default.
• W4221130850 countsByYear W42211308502022 @default.
• W4221130850 countsByYear W42211308502023 @default.
• W4221130850 crossrefType "journal-article" @default.
• W4221130850 hasAuthorship W4221130850A5023793169 @default.
• W4221130850 hasAuthorship W4221130850A5044602576 @default.
• W4221130850 hasAuthorship W4221130850A5061420652 @default.
• W4221130850 hasConcept C124504099 @default.
• W4221130850 hasConcept C134306372 @default.
• W4221130850 hasConcept C153180895 @default.
• W4221130850 hasConcept C154945302 @default.
• W4221130850 hasConcept C177148314 @default.
• W4221130850 hasConcept C31601959 @default.
• W4221130850 hasConcept C31972630 @default.
• W4221130850 hasConcept C33923547 @default.
• W4221130850 hasConcept C41008148 @default.
• W4221130850 hasConcept C89600930 @default.
• W4221130850 hasConceptScore W4221130850C124504099 @default.
• W4221130850 hasConceptScore W4221130850C134306372 @default.
• W4221130850 hasConceptScore W4221130850C153180895 @default.
• W4221130850 hasConceptScore W4221130850C154945302 @default.
• W4221130850 hasConceptScore W4221130850C177148314 @default.
• W4221130850 hasConceptScore W4221130850C31601959 @default.
• W4221130850 hasConceptScore W4221130850C31972630 @default.
• W4221130850 hasConceptScore W4221130850C33923547 @default.
• W4221130850 hasConceptScore W4221130850C41008148 @default.
• W4221130850 hasConceptScore W4221130850C89600930 @default.
• W4221130850 hasFunder F4320322120 @default.
• W4221130850 hasLocation W42211308501 @default.
• W4221130850 hasLocation W42211308502 @default.
• W4221130850 hasOpenAccess W4221130850 @default.
• W4221130850 hasPrimaryLocation W42211308501 @default.
• W4221130850 hasRelatedWork W1507266234 @default.
• W4221130850 hasRelatedWork W1669643531 @default.
• W4221130850 hasRelatedWork W2110230079 @default.
• W4221130850 hasRelatedWork W2117664411 @default.
• W4221130850 hasRelatedWork W2117933325 @default.
• W4221130850 hasRelatedWork W2122581818 @default.
• W4221130850 hasRelatedWork W2159066190 @default.
• W4221130850 hasRelatedWork W2549936415 @default.
• W4221130850 hasRelatedWork W2739874619 @default.
• W4221130850 hasRelatedWork W1967061043 @default.
• W4221130850 hasVolume "145" @default.
• W4221130850 isParatext "false" @default.
• W4221130850 isRetracted "false" @default.
• W4221130850 workType "article" @default.