FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3187844553> ?p ?o ?g. }

• W3187844553 endingPage "102136" @default.
• W3187844553 startingPage "102136" @default.
• W3187844553 abstract "Developing deep learning models to analyze histology images has been computationally challenging, as the massive size of the images causes excessive strain on all parts of the computing pipeline. This paper proposes a novel deep learning-based methodology for improving the computational efficiency of histology image classification. The proposed approach is robust when used with images that have reduced input resolution, and it can be trained effectively with limited labeled data. Moreover, our approach operates at either the tissue- or slide-level, removing the need for laborious patch-level labeling. Our method uses knowledge distillation to transfer knowledge from a teacher model pre-trained at high resolution to a student model trained on the same images at a considerably lower resolution. Also, to address the lack of large-scale labeled histology image datasets, we perform the knowledge distillation in a self-supervised fashion. We evaluate our approach on three distinct histology image datasets associated with celiac disease, lung adenocarcinoma, and renal cell carcinoma. Our results on these datasets demonstrate that a combination of knowledge distillation and self-supervision allows the student model to approach and, in some cases, surpass the teacher model's classification accuracy while being much more computationally efficient. Additionally, we observe an increase in student classification performance as the size of the unlabeled dataset increases, indicating that there is potential for this method to scale further with additional unlabeled data. Our model outperforms the high-resolution teacher model for celiac disease in accuracy, F1-score, precision, and recall while requiring 4 times fewer computations. For lung adenocarcinoma, our results at 1.25× magnification are within 1.5% of the results for the teacher model at 10× magnification, with a reduction in computational cost by a factor of 64. Our model on renal cell carcinoma at 1.25× magnification performs within 1% of the teacher model at 5× magnification while requiring 16 times fewer computations. Furthermore, our celiac disease outcomes benefit from additional performance scaling with the use of more unlabeled data. In the case of 0.625× magnification, using unlabeled data improves accuracy by 4% over the tissue-level baseline. Therefore, our approach can improve the feasibility of deep learning solutions for digital pathology on standard computational hardware and infrastructures." @default.
• W3187844553 created "2021-08-16" @default.
• W3187844553 creator A5001118529 @default.
• W3187844553 creator A5040432994 @default.
• W3187844553 creator A5051476268 @default.
• W3187844553 creator A5080286505 @default.
• W3187844553 creator A5082736347 @default.
• W3187844553 date "2021-09-01" @default.
• W3187844553 modified "2023-10-16" @default.
• W3187844553 title "Resolution-based distillation for efficient histology image classification" @default.
• W3187844553 cites W1173721425 @default.
• W3187844553 cites W1964055730 @default.
• W3187844553 cites W1965555277 @default.
• W3187844553 cites W1986348058 @default.
• W3187844553 cites W2010596067 @default.
• W3187844553 cites W2046560792 @default.
• W3187844553 cites W2049674541 @default.
• W3187844553 cites W2063759711 @default.
• W3187844553 cites W2158106117 @default.
• W3187844553 cites W2211456655 @default.
• W3187844553 cites W2251438188 @default.
• W3187844553 cites W2280957887 @default.
• W3187844553 cites W2594760301 @default.
• W3187844553 cites W2621015632 @default.
• W3187844553 cites W2726542547 @default.
• W3187844553 cites W2762672048 @default.
• W3187844553 cites W2796207353 @default.
• W3187844553 cites W2808343649 @default.
• W3187844553 cites W2900654702 @default.
• W3187844553 cites W2901505625 @default.
• W3187844553 cites W2907462222 @default.
• W3187844553 cites W2913326971 @default.
• W3187844553 cites W2937006088 @default.
• W3187844553 cites W2954719570 @default.
• W3187844553 cites W2956228567 @default.
• W3187844553 cites W2964345665 @default.
• W3187844553 cites W2981841914 @default.
• W3187844553 cites W3046446892 @default.
• W3187844553 cites W3082947597 @default.
• W3187844553 cites W3100084586 @default.
• W3187844553 cites W3128210037 @default.
• W3187844553 cites W3138848412 @default.
• W3187844553 cites W3142933006 @default.
• W3187844553 cites W4236836788 @default.
• W3187844553 cites W4294358748 @default.
• W3187844553 doi "https://doi.org/10.1016/j.artmed.2021.102136" @default.
• W3187844553 hasPubMedCentralId "https://www.ncbi.nlm.nih.gov/pmc/articles/8449014" @default.
• W3187844553 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/34531005" @default.
• W3187844553 hasPublicationYear "2021" @default.
• W3187844553 type Work @default.
• W3187844553 sameAs 3187844553 @default.
• W3187844553 citedByCount "12" @default.
• W3187844553 countsByYear W31878445532021 @default.
• W3187844553 countsByYear W31878445532022 @default.
• W3187844553 countsByYear W31878445532023 @default.
• W3187844553 crossrefType "journal-article" @default.
• W3187844553 hasAuthorship W3187844553A5001118529 @default.
• W3187844553 hasAuthorship W3187844553A5040432994 @default.
• W3187844553 hasAuthorship W3187844553A5051476268 @default.
• W3187844553 hasAuthorship W3187844553A5080286505 @default.
• W3187844553 hasAuthorship W3187844553A5082736347 @default.
• W3187844553 hasBestOaLocation W31878445531 @default.
• W3187844553 hasConcept C108583219 @default.
• W3187844553 hasConcept C119857082 @default.
• W3187844553 hasConcept C121332964 @default.
• W3187844553 hasConcept C150899416 @default.
• W3187844553 hasConcept C153180895 @default.
• W3187844553 hasConcept C154945302 @default.
• W3187844553 hasConcept C178790620 @default.
• W3187844553 hasConcept C185592680 @default.
• W3187844553 hasConcept C199360897 @default.
• W3187844553 hasConcept C204030448 @default.
• W3187844553 hasConcept C2778755073 @default.
• W3187844553 hasConcept C41008148 @default.
• W3187844553 hasConcept C43521106 @default.
• W3187844553 hasConcept C62520636 @default.
• W3187844553 hasConcept C81669768 @default.
• W3187844553 hasConceptScore W3187844553C108583219 @default.
• W3187844553 hasConceptScore W3187844553C119857082 @default.
• W3187844553 hasConceptScore W3187844553C121332964 @default.
• W3187844553 hasConceptScore W3187844553C150899416 @default.
• W3187844553 hasConceptScore W3187844553C153180895 @default.
• W3187844553 hasConceptScore W3187844553C154945302 @default.
• W3187844553 hasConceptScore W3187844553C178790620 @default.
• W3187844553 hasConceptScore W3187844553C185592680 @default.
• W3187844553 hasConceptScore W3187844553C199360897 @default.
• W3187844553 hasConceptScore W3187844553C204030448 @default.
• W3187844553 hasConceptScore W3187844553C2778755073 @default.
• W3187844553 hasConceptScore W3187844553C41008148 @default.
• W3187844553 hasConceptScore W3187844553C43521106 @default.
• W3187844553 hasConceptScore W3187844553C62520636 @default.
• W3187844553 hasConceptScore W3187844553C81669768 @default.
• W3187844553 hasFunder F4320337351 @default.
• W3187844553 hasFunder F4320337372 @default.
• W3187844553 hasLocation W31878445531 @default.
• W3187844553 hasLocation W31878445532 @default.
• W3187844553 hasLocation W31878445533 @default.
• W3187844553 hasLocation W31878445534 @default.

• next