Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386088139> ?p ?o ?g. }
Showing items 1 to 86 of
86
with 100 items per page.
- W4386088139 endingPage "1" @default.
- W4386088139 startingPage "1" @default.
- W4386088139 abstract "Convolutional neural networks (CNNs) are a promising technique for automated glaucoma diagnosis from images of the fundus, and these images are routinely acquired as part of an ophthalmic exam. Nevertheless, CNNs typically require a large amount of well-labeled data for training, which may not be available in many biomedical image classification applications, especially when diseases are rare and where labeling by experts is costly. This paper makes two contributions to address this issue: (1) It extends the conventional Siamese network and introduces a training method for low-shot learning when labeled data are limited and imbalanced, and (2) it introduces a novel semi-supervised learning strategy that uses additional unlabeled training data to achieve greater accuracy. Our proposed multi-task Siamese network (MTSN) can employ any backbone CNN, and we demonstrate with four backbone CNNs that its accuracy with limited training data approaches the accuracy of backbone CNNs trained with a dataset that is 50 times larger. We also introduce One-Vote Veto (OVV) self-training, a semi-supervised learning strategy that is designed specifically for MTSNs. By taking both self-predictions and contrastive predictions of the unlabeled training data into account, OVV self-training provides additional pseudo labels for fine-tuning a pre-trained MTSN. Using a large (imbalanced) dataset with 66,715 fundus photographs acquired over 15 years, extensive experimental results demonstrate the effectiveness of low-shot learning with MTSN and semi-supervised learning with OVV self-training. Three additional, smaller clinical datasets of fundus images acquired under different conditions (cameras, instruments, locations, populations) are used to demonstrate the generalizability of the proposed methods." @default.
- W4386088139 created "2023-08-24" @default.
- W4386088139 creator A5009959838 @default.
- W4386088139 creator A5016855297 @default.
- W4386088139 creator A5038867899 @default.
- W4386088139 creator A5040564869 @default.
- W4386088139 creator A5043994887 @default.
- W4386088139 creator A5063018832 @default.
- W4386088139 creator A5063363976 @default.
- W4386088139 date "2023-01-01" @default.
- W4386088139 modified "2023-10-15" @default.
- W4386088139 title "One-Vote Veto: Semi-Supervised Learning for Low-Shot Glaucoma Diagnosis" @default.
- W4386088139 doi "https://doi.org/10.1109/tmi.2023.3307689" @default.
- W4386088139 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/37610903" @default.
- W4386088139 hasPublicationYear "2023" @default.
- W4386088139 type Work @default.
- W4386088139 citedByCount "0" @default.
- W4386088139 crossrefType "journal-article" @default.
- W4386088139 hasAuthorship W4386088139A5009959838 @default.
- W4386088139 hasAuthorship W4386088139A5016855297 @default.
- W4386088139 hasAuthorship W4386088139A5038867899 @default.
- W4386088139 hasAuthorship W4386088139A5040564869 @default.
- W4386088139 hasAuthorship W4386088139A5043994887 @default.
- W4386088139 hasAuthorship W4386088139A5063018832 @default.
- W4386088139 hasAuthorship W4386088139A5063363976 @default.
- W4386088139 hasBestOaLocation W43860881391 @default.
- W4386088139 hasConcept C108583219 @default.
- W4386088139 hasConcept C118487528 @default.
- W4386088139 hasConcept C119857082 @default.
- W4386088139 hasConcept C127413603 @default.
- W4386088139 hasConcept C136389625 @default.
- W4386088139 hasConcept C153180895 @default.
- W4386088139 hasConcept C154945302 @default.
- W4386088139 hasConcept C162324750 @default.
- W4386088139 hasConcept C187736073 @default.
- W4386088139 hasConcept C2776145971 @default.
- W4386088139 hasConcept C2776391266 @default.
- W4386088139 hasConcept C2780451532 @default.
- W4386088139 hasConcept C2992734406 @default.
- W4386088139 hasConcept C31972630 @default.
- W4386088139 hasConcept C41008148 @default.
- W4386088139 hasConcept C50644808 @default.
- W4386088139 hasConcept C71924100 @default.
- W4386088139 hasConcept C78519656 @default.
- W4386088139 hasConcept C81363708 @default.
- W4386088139 hasConceptScore W4386088139C108583219 @default.
- W4386088139 hasConceptScore W4386088139C118487528 @default.
- W4386088139 hasConceptScore W4386088139C119857082 @default.
- W4386088139 hasConceptScore W4386088139C127413603 @default.
- W4386088139 hasConceptScore W4386088139C136389625 @default.
- W4386088139 hasConceptScore W4386088139C153180895 @default.
- W4386088139 hasConceptScore W4386088139C154945302 @default.
- W4386088139 hasConceptScore W4386088139C162324750 @default.
- W4386088139 hasConceptScore W4386088139C187736073 @default.
- W4386088139 hasConceptScore W4386088139C2776145971 @default.
- W4386088139 hasConceptScore W4386088139C2776391266 @default.
- W4386088139 hasConceptScore W4386088139C2780451532 @default.
- W4386088139 hasConceptScore W4386088139C2992734406 @default.
- W4386088139 hasConceptScore W4386088139C31972630 @default.
- W4386088139 hasConceptScore W4386088139C41008148 @default.
- W4386088139 hasConceptScore W4386088139C50644808 @default.
- W4386088139 hasConceptScore W4386088139C71924100 @default.
- W4386088139 hasConceptScore W4386088139C78519656 @default.
- W4386088139 hasConceptScore W4386088139C81363708 @default.
- W4386088139 hasFunder F4320332161 @default.
- W4386088139 hasFunder F4320335787 @default.
- W4386088139 hasFunder F4320337350 @default.
- W4386088139 hasLocation W43860881391 @default.
- W4386088139 hasLocation W43860881392 @default.
- W4386088139 hasOpenAccess W4386088139 @default.
- W4386088139 hasPrimaryLocation W43860881391 @default.
- W4386088139 hasRelatedWork W2731899572 @default.
- W4386088139 hasRelatedWork W3116150086 @default.
- W4386088139 hasRelatedWork W3133861977 @default.
- W4386088139 hasRelatedWork W3192794374 @default.
- W4386088139 hasRelatedWork W4200173597 @default.
- W4386088139 hasRelatedWork W4220686584 @default.
- W4386088139 hasRelatedWork W4246751904 @default.
- W4386088139 hasRelatedWork W4310880831 @default.
- W4386088139 hasRelatedWork W4312417841 @default.
- W4386088139 hasRelatedWork W4321369474 @default.
- W4386088139 isParatext "false" @default.
- W4386088139 isRetracted "false" @default.
- W4386088139 workType "article" @default.