Matches in SemOpenAlex for { <https://semopenalex.org/work/W3204337192> ?p ?o ?g. }
Showing items 1 to 88 of
88
with 100 items per page.
- W3204337192 abstract "PURPOSE: Deployment of deep models for clinical decision making should not only provide predicted outcomes, but also insights on how decisions are made. Considering the interpretability of Transformer models, and the power of graph networks in analyzing the inherent hierarchy of biological signals, a combined approach would be the next generation solution in computer aided interventions. In this study, we propose a framework for classification and visualization of surgical mass spectrometry data using Graph Transformer model to empower the interpretability of breast surgical margin assessment. METHODS: Using the iKnife, 144 burns (103 normal, 41 cancer) were collected and converted to multi-level graph structures. A Graph Transformer model was modified to output the intermediate attention parameters of the network. Beside ablation and prospective study, we propose multiple attention visualization approaches to facilitate the interpretability. RESULTS: In a 4-fold cross validation experiment, an average classification AUC of 95.6% was achieved, outperforming baseline models. We could also distinguish and visualize clear pattern of attention difference between burns. For instance, cancerous and normal burns gather more attention in the lower and higher subbands of the spectra respectively. Looking at cancer subtype prospectively, a pattern of cancer progression was also observed in the attention features. CONCLUSION: Graph Transformers are powerful in providing high network interpretability. When paired with proper visualization, they can be deployed for computer assisted interventions." @default.
- W3204337192 created "2021-10-11" @default.
- W3204337192 creator A5002749573 @default.
- W3204337192 creator A5008717552 @default.
- W3204337192 creator A5022874103 @default.
- W3204337192 creator A5026941433 @default.
- W3204337192 creator A5038634249 @default.
- W3204337192 creator A5040401197 @default.
- W3204337192 creator A5041571660 @default.
- W3204337192 creator A5065285863 @default.
- W3204337192 creator A5074148873 @default.
- W3204337192 creator A5075090998 @default.
- W3204337192 creator A5078370200 @default.
- W3204337192 creator A5079682225 @default.
- W3204337192 creator A5084753910 @default.
- W3204337192 date "2021-01-01" @default.
- W3204337192 modified "2023-10-18" @default.
- W3204337192 title "Graph Transformers for Characterization and Interpretation of Surgical Margins" @default.
- W3204337192 cites W2014058353 @default.
- W3204337192 cites W2505592464 @default.
- W3204337192 cites W2987105127 @default.
- W3204337192 cites W2989491368 @default.
- W3204337192 cites W2999610835 @default.
- W3204337192 cites W3035486570 @default.
- W3204337192 cites W3036019667 @default.
- W3204337192 cites W3116286104 @default.
- W3204337192 cites W3131844474 @default.
- W3204337192 cites W3159603107 @default.
- W3204337192 doi "https://doi.org/10.1007/978-3-030-87234-2_9" @default.
- W3204337192 hasPublicationYear "2021" @default.
- W3204337192 type Work @default.
- W3204337192 sameAs 3204337192 @default.
- W3204337192 citedByCount "1" @default.
- W3204337192 crossrefType "book-chapter" @default.
- W3204337192 hasAuthorship W3204337192A5002749573 @default.
- W3204337192 hasAuthorship W3204337192A5008717552 @default.
- W3204337192 hasAuthorship W3204337192A5022874103 @default.
- W3204337192 hasAuthorship W3204337192A5026941433 @default.
- W3204337192 hasAuthorship W3204337192A5038634249 @default.
- W3204337192 hasAuthorship W3204337192A5040401197 @default.
- W3204337192 hasAuthorship W3204337192A5041571660 @default.
- W3204337192 hasAuthorship W3204337192A5065285863 @default.
- W3204337192 hasAuthorship W3204337192A5074148873 @default.
- W3204337192 hasAuthorship W3204337192A5075090998 @default.
- W3204337192 hasAuthorship W3204337192A5078370200 @default.
- W3204337192 hasAuthorship W3204337192A5079682225 @default.
- W3204337192 hasAuthorship W3204337192A5084753910 @default.
- W3204337192 hasConcept C119857082 @default.
- W3204337192 hasConcept C121332964 @default.
- W3204337192 hasConcept C124101348 @default.
- W3204337192 hasConcept C132525143 @default.
- W3204337192 hasConcept C154945302 @default.
- W3204337192 hasConcept C165801399 @default.
- W3204337192 hasConcept C2781067378 @default.
- W3204337192 hasConcept C36464697 @default.
- W3204337192 hasConcept C41008148 @default.
- W3204337192 hasConcept C62520636 @default.
- W3204337192 hasConcept C66322947 @default.
- W3204337192 hasConcept C80444323 @default.
- W3204337192 hasConceptScore W3204337192C119857082 @default.
- W3204337192 hasConceptScore W3204337192C121332964 @default.
- W3204337192 hasConceptScore W3204337192C124101348 @default.
- W3204337192 hasConceptScore W3204337192C132525143 @default.
- W3204337192 hasConceptScore W3204337192C154945302 @default.
- W3204337192 hasConceptScore W3204337192C165801399 @default.
- W3204337192 hasConceptScore W3204337192C2781067378 @default.
- W3204337192 hasConceptScore W3204337192C36464697 @default.
- W3204337192 hasConceptScore W3204337192C41008148 @default.
- W3204337192 hasConceptScore W3204337192C62520636 @default.
- W3204337192 hasConceptScore W3204337192C66322947 @default.
- W3204337192 hasConceptScore W3204337192C80444323 @default.
- W3204337192 hasLocation W32043371921 @default.
- W3204337192 hasOpenAccess W3204337192 @default.
- W3204337192 hasPrimaryLocation W32043371921 @default.
- W3204337192 hasRelatedWork W10015831 @default.
- W3204337192 hasRelatedWork W10931660 @default.
- W3204337192 hasRelatedWork W13464624 @default.
- W3204337192 hasRelatedWork W2076915 @default.
- W3204337192 hasRelatedWork W2701911 @default.
- W3204337192 hasRelatedWork W3989876 @default.
- W3204337192 hasRelatedWork W5005399 @default.
- W3204337192 hasRelatedWork W7002624 @default.
- W3204337192 hasRelatedWork W8190784 @default.
- W3204337192 hasRelatedWork W2104954 @default.
- W3204337192 isParatext "false" @default.
- W3204337192 isRetracted "false" @default.
- W3204337192 magId "3204337192" @default.
- W3204337192 workType "book-chapter" @default.