Matches in SemOpenAlex for { <https://semopenalex.org/work/W4320060316> ?p ?o ?g. }
Showing items 1 to 84 of
84
with 100 items per page.
- W4320060316 abstract "AI-powered Medical Imaging has recently achieved enormous attention due to its ability to provide fast-paced healthcare diagnoses. However, it usually suffers from a lack of high-quality datasets due to high annotation cost, interobserver variability, human annotator error, and errors in computer-generated labels. Deep learning models trained on noisy labelled datasets are sensitive to the noise type and lead to less generalization on the unseen samples. To address this challenge, we propose a Robust Stochastic Knowledge Distillation (RoS-KD) framework which mimics the notion of learning a topic from multiple sources to ensure deterrence in learning noisy information. More specifically, RoS-KD learns a smooth, well-informed, and robust student manifold by distilling knowledge from multiple teachers trained on overlapping subsets of training data. Our extensive experiments on popular medical imaging classification tasks (cardiopulmonary disease and lesion classification) using real-world datasets, show the performance benefit of RoS-KD, its ability to distill knowledge from many popular large networks (ResNet-50, DenseNet-121, MobileNetV2) in a comparatively small network, and its robustness to adversarial attacks (PGD, FSGM). More specifically, RoS-KD achieves >2% and > 4% improvement on F1-score for lesion classification and cardiopulmonary disease classification tasks, respectively, when the underlying student is ResNet-18 against recent competitive knowledge distillation baseline. Additionally, on cardiopulmonary disease classification task, RoS-KD outperforms most of the SOTA baselines by ~1% gain in AUC score." @default.
- W4320060316 created "2023-02-12" @default.
- W4320060316 creator A5022034850 @default.
- W4320060316 creator A5026263464 @default.
- W4320060316 creator A5033862822 @default.
- W4320060316 creator A5034136464 @default.
- W4320060316 creator A5047170063 @default.
- W4320060316 creator A5048522863 @default.
- W4320060316 date "2022-11-01" @default.
- W4320060316 modified "2023-09-26" @default.
- W4320060316 title "RoS-KD: A Robust Stochastic Knowledge Distillation Approach for Noisy Medical Imaging" @default.
- W4320060316 cites W2183341477 @default.
- W4320060316 cites W2194775991 @default.
- W4320060316 cites W2747909401 @default.
- W4320060316 cites W2937408455 @default.
- W4320060316 cites W2962708065 @default.
- W4320060316 cites W2962975664 @default.
- W4320060316 cites W2963967185 @default.
- W4320060316 cites W2993044507 @default.
- W4320060316 cites W2995225687 @default.
- W4320060316 cites W2996834012 @default.
- W4320060316 cites W3101156210 @default.
- W4320060316 cites W3110181119 @default.
- W4320060316 cites W3110846353 @default.
- W4320060316 cites W3177694721 @default.
- W4320060316 cites W3209663319 @default.
- W4320060316 cites W4226050456 @default.
- W4320060316 doi "https://doi.org/10.1109/icdm54844.2022.00118" @default.
- W4320060316 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/37038389" @default.
- W4320060316 hasPublicationYear "2022" @default.
- W4320060316 type Work @default.
- W4320060316 citedByCount "2" @default.
- W4320060316 countsByYear W43200603162023 @default.
- W4320060316 crossrefType "proceedings-article" @default.
- W4320060316 hasAuthorship W4320060316A5022034850 @default.
- W4320060316 hasAuthorship W4320060316A5026263464 @default.
- W4320060316 hasAuthorship W4320060316A5033862822 @default.
- W4320060316 hasAuthorship W4320060316A5034136464 @default.
- W4320060316 hasAuthorship W4320060316A5047170063 @default.
- W4320060316 hasAuthorship W4320060316A5048522863 @default.
- W4320060316 hasBestOaLocation W43200603162 @default.
- W4320060316 hasConcept C104317684 @default.
- W4320060316 hasConcept C108583219 @default.
- W4320060316 hasConcept C119857082 @default.
- W4320060316 hasConcept C142724271 @default.
- W4320060316 hasConcept C153180895 @default.
- W4320060316 hasConcept C154945302 @default.
- W4320060316 hasConcept C185592680 @default.
- W4320060316 hasConcept C41008148 @default.
- W4320060316 hasConcept C534262118 @default.
- W4320060316 hasConcept C55493867 @default.
- W4320060316 hasConcept C63479239 @default.
- W4320060316 hasConcept C71924100 @default.
- W4320060316 hasConceptScore W4320060316C104317684 @default.
- W4320060316 hasConceptScore W4320060316C108583219 @default.
- W4320060316 hasConceptScore W4320060316C119857082 @default.
- W4320060316 hasConceptScore W4320060316C142724271 @default.
- W4320060316 hasConceptScore W4320060316C153180895 @default.
- W4320060316 hasConceptScore W4320060316C154945302 @default.
- W4320060316 hasConceptScore W4320060316C185592680 @default.
- W4320060316 hasConceptScore W4320060316C41008148 @default.
- W4320060316 hasConceptScore W4320060316C534262118 @default.
- W4320060316 hasConceptScore W4320060316C55493867 @default.
- W4320060316 hasConceptScore W4320060316C63479239 @default.
- W4320060316 hasConceptScore W4320060316C71924100 @default.
- W4320060316 hasLocation W43200603161 @default.
- W4320060316 hasLocation W43200603162 @default.
- W4320060316 hasLocation W43200603163 @default.
- W4320060316 hasLocation W43200603164 @default.
- W4320060316 hasOpenAccess W4320060316 @default.
- W4320060316 hasPrimaryLocation W43200603161 @default.
- W4320060316 hasRelatedWork W2773120646 @default.
- W4320060316 hasRelatedWork W3014300295 @default.
- W4320060316 hasRelatedWork W3164822677 @default.
- W4320060316 hasRelatedWork W4223943233 @default.
- W4320060316 hasRelatedWork W4225161397 @default.
- W4320060316 hasRelatedWork W4250304930 @default.
- W4320060316 hasRelatedWork W4309045103 @default.
- W4320060316 hasRelatedWork W4309637067 @default.
- W4320060316 hasRelatedWork W4312200629 @default.
- W4320060316 hasRelatedWork W4360585206 @default.
- W4320060316 isParatext "false" @default.
- W4320060316 isRetracted "false" @default.
- W4320060316 workType "article" @default.