FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3210158061> ?p ?o ?g. }

• W3210158061 abstract "Semi-supervised learning (SSL) is a promising machine learning paradigm to address the issue of label scarcity in medical imaging. SSL methods were originally developed in image classification. The state-of-the-art SSL methods in image classification utilise consistency regularisation to learn unlabelled predictions which are invariant to input level perturbations. However, image level perturbations violate the cluster assumption in the setting of segmentation. Moreover, existing image level perturbations are hand-crafted which could be sub-optimal. Therefore, it is a not trivial to straightforwardly adapt existing SSL image classification methods in segmentation. In this paper, we propose MisMatch, a semi-supervised segmentation framework based on the consistency between paired predictions which are derived from two differently learnt morphological feature perturbations. MisMatch consists of an encoder and two decoders. One decoder learns positive attention for foreground on unlabelled data thereby generating dilated features of foreground. The other decoder learns negative attention for foreground on the same unlabelled data thereby generating eroded features of foreground. We first develop a 2D U-net based MisMatch framework and perform extensive cross-validation on a CT-based pulmonary vessel segmentation task and show that MisMatch statistically outperforms state-of-the-art semi-supervised methods when only 6.25% of the total labels are used. In a second experiment, we show that U-net based MisMatch outperforms state-of-the-art methods on an MRI-based brain tumour segmentation task. In a third experiment, we show that a 3D MisMatch outperforms a previous method using input level augmentations, on a left atrium segmentation task. Lastly, we find that the performance improvement of MisMatch over the baseline might originate from its better calibration." @default.
• W3210158061 created "2021-11-08" @default.
• W3210158061 creator A5007350183 @default.
• W3210158061 creator A5009829456 @default.
• W3210158061 creator A5010770082 @default.
• W3210158061 creator A5040477021 @default.
• W3210158061 creator A5042005913 @default.
• W3210158061 creator A5073094674 @default.
• W3210158061 creator A5075265609 @default.
• W3210158061 creator A5087083706 @default.
• W3210158061 creator A5088669096 @default.
• W3210158061 date "2021-10-23" @default.
• W3210158061 modified "2023-10-18" @default.
• W3210158061 title "MisMatch: Calibrated Segmentation via Consistency on Differential Morphological Feature Perturbations with Limited Labels" @default.
• W3210158061 cites W1599263113 @default.
• W3210158061 cites W1641498739 @default.
• W3210158061 cites W2108501770 @default.
• W3210158061 cites W2145494108 @default.
• W3210158061 cites W2213701027 @default.
• W3210158061 cites W2556967412 @default.
• W3210158061 cites W2592691248 @default.
• W3210158061 cites W2798376494 @default.
• W3210158061 cites W2909869271 @default.
• W3210158061 cites W2949284809 @default.
• W3210158061 cites W2949416428 @default.
• W3210158061 cites W2962736273 @default.
• W3210158061 cites W2962914239 @default.
• W3210158061 cites W2963410064 @default.
• W3210158061 cites W2963956526 @default.
• W3210158061 cites W2964121744 @default.
• W3210158061 cites W2964159205 @default.
• W3210158061 cites W2964212410 @default.
• W3210158061 cites W2970971581 @default.
• W3210158061 cites W2979632994 @default.
• W3210158061 cites W2979651795 @default.
• W3210158061 cites W2992308087 @default.
• W3210158061 cites W2996501936 @default.
• W3210158061 cites W3035680157 @default.
• W3210158061 cites W3091067756 @default.
• W3210158061 cites W3091785623 @default.
• W3210158061 cites W3092515963 @default.
• W3210158061 cites W3096165964 @default.
• W3210158061 cites W3099570996 @default.
• W3210158061 cites W3102631365 @default.
• W3210158061 cites W3107695429 @default.
• W3210158061 cites W3127064459 @default.
• W3210158061 cites W3127902700 @default.
• W3210158061 doi "https://doi.org/10.48550/arxiv.2110.12179" @default.
• W3210158061 hasPublicationYear "2021" @default.
• W3210158061 type Work @default.
• W3210158061 sameAs 3210158061 @default.
• W3210158061 citedByCount "0" @default.
• W3210158061 crossrefType "posted-content" @default.
• W3210158061 hasAuthorship W3210158061A5007350183 @default.
• W3210158061 hasAuthorship W3210158061A5009829456 @default.
• W3210158061 hasAuthorship W3210158061A5010770082 @default.
• W3210158061 hasAuthorship W3210158061A5040477021 @default.
• W3210158061 hasAuthorship W3210158061A5042005913 @default.
• W3210158061 hasAuthorship W3210158061A5073094674 @default.
• W3210158061 hasAuthorship W3210158061A5075265609 @default.
• W3210158061 hasAuthorship W3210158061A5087083706 @default.
• W3210158061 hasAuthorship W3210158061A5088669096 @default.
• W3210158061 hasBestOaLocation W32101580611 @default.
• W3210158061 hasConcept C111919701 @default.
• W3210158061 hasConcept C118505674 @default.
• W3210158061 hasConcept C124504099 @default.
• W3210158061 hasConcept C138885662 @default.
• W3210158061 hasConcept C153180895 @default.
• W3210158061 hasConcept C154945302 @default.
• W3210158061 hasConcept C2776401178 @default.
• W3210158061 hasConcept C2776436953 @default.
• W3210158061 hasConcept C41008148 @default.
• W3210158061 hasConcept C41895202 @default.
• W3210158061 hasConcept C65885262 @default.
• W3210158061 hasConcept C89600930 @default.
• W3210158061 hasConceptScore W3210158061C111919701 @default.
• W3210158061 hasConceptScore W3210158061C118505674 @default.
• W3210158061 hasConceptScore W3210158061C124504099 @default.
• W3210158061 hasConceptScore W3210158061C138885662 @default.
• W3210158061 hasConceptScore W3210158061C153180895 @default.
• W3210158061 hasConceptScore W3210158061C154945302 @default.
• W3210158061 hasConceptScore W3210158061C2776401178 @default.
• W3210158061 hasConceptScore W3210158061C2776436953 @default.
• W3210158061 hasConceptScore W3210158061C41008148 @default.
• W3210158061 hasConceptScore W3210158061C41895202 @default.
• W3210158061 hasConceptScore W3210158061C65885262 @default.
• W3210158061 hasConceptScore W3210158061C89600930 @default.
• W3210158061 hasLocation W32101580611 @default.
• W3210158061 hasOpenAccess W3210158061 @default.
• W3210158061 hasPrimaryLocation W32101580611 @default.
• W3210158061 hasRelatedWork W1507687735 @default.
• W3210158061 hasRelatedWork W1631910785 @default.
• W3210158061 hasRelatedWork W1974884835 @default.
• W3210158061 hasRelatedWork W2005476934 @default.
• W3210158061 hasRelatedWork W2150405159 @default.
• W3210158061 hasRelatedWork W2344532017 @default.
• W3210158061 hasRelatedWork W2897195263 @default.
• W3210158061 hasRelatedWork W2903115243 @default.
• W3210158061 hasRelatedWork W2948522034 @default.
• W3210158061 hasRelatedWork W4361265312 @default.

• next