Matches in SemOpenAlex for { <https://semopenalex.org/work/W2807515278> ?p ?o ?g. }
Showing items 1 to 77 of
77
with 100 items per page.
- W2807515278 abstract "The convolutional neural network (CNN) has become a powerful tool for various biomedical image analysis tasks, but there is a lack of visual explanation for the machinery of CNNs. In this paper, we present a novel algorithm, Respond-weighted Class Activation Mapping (Respond-CAM), for making CNN-based models interpretable by visualizing input regions that are important for predictions, especially for biomedical 3D imaging data inputs. Our method uses the gradients of any target concept (e.g. the score of target class) that flows into a convolutional layer. The weighted feature maps are combined to produce a heatmap that highlights the important regions in the image for predicting the target concept. We prove a preferable sum-to-score property of the Respond-CAM and verify its significant improvement on 3D images from the current state-of-the-art approach. Our tests on Cellular Electron Cryo-Tomography 3D images show that Respond-CAM achieves superior performance on visualizing the CNNs with 3D biomedical images inputs, and is able to get reasonably good results on visualizing the CNNs with natural image inputs. The Respond-CAM is an efficient and reliable approach for visualizing the CNN machinery, and is applicable to a wide variety of CNN model families and image analysis tasks." @default.
- W2807515278 created "2018-06-13" @default.
- W2807515278 creator A5000807861 @default.
- W2807515278 creator A5035154955 @default.
- W2807515278 creator A5060390157 @default.
- W2807515278 creator A5072994598 @default.
- W2807515278 creator A5085149886 @default.
- W2807515278 date "2018-05-31" @default.
- W2807515278 modified "2023-09-26" @default.
- W2807515278 title "Respond-CAM: Analyzing Deep Models for 3D Imaging Data by Visualizations" @default.
- W2807515278 cites W1825675169 @default.
- W2807515278 cites W1849277567 @default.
- W2807515278 cites W1915485278 @default.
- W2807515278 cites W1985514943 @default.
- W2807515278 cites W2061835758 @default.
- W2807515278 cites W2123045220 @default.
- W2807515278 cites W2295107390 @default.
- W2807515278 cites W2591330651 @default.
- W2807515278 cites W2962851944 @default.
- W2807515278 doi "https://doi.org/10.48550/arxiv.1806.00102" @default.
- W2807515278 hasPublicationYear "2018" @default.
- W2807515278 type Work @default.
- W2807515278 sameAs 2807515278 @default.
- W2807515278 citedByCount "2" @default.
- W2807515278 countsByYear W28075152782019 @default.
- W2807515278 crossrefType "posted-content" @default.
- W2807515278 hasAuthorship W2807515278A5000807861 @default.
- W2807515278 hasAuthorship W2807515278A5035154955 @default.
- W2807515278 hasAuthorship W2807515278A5060390157 @default.
- W2807515278 hasAuthorship W2807515278A5072994598 @default.
- W2807515278 hasAuthorship W2807515278A5085149886 @default.
- W2807515278 hasBestOaLocation W28075152781 @default.
- W2807515278 hasConcept C108583219 @default.
- W2807515278 hasConcept C111472728 @default.
- W2807515278 hasConcept C115961682 @default.
- W2807515278 hasConcept C119857082 @default.
- W2807515278 hasConcept C138885662 @default.
- W2807515278 hasConcept C153180895 @default.
- W2807515278 hasConcept C154945302 @default.
- W2807515278 hasConcept C189950617 @default.
- W2807515278 hasConcept C2776401178 @default.
- W2807515278 hasConcept C2777212361 @default.
- W2807515278 hasConcept C36464697 @default.
- W2807515278 hasConcept C41008148 @default.
- W2807515278 hasConcept C41895202 @default.
- W2807515278 hasConcept C81363708 @default.
- W2807515278 hasConceptScore W2807515278C108583219 @default.
- W2807515278 hasConceptScore W2807515278C111472728 @default.
- W2807515278 hasConceptScore W2807515278C115961682 @default.
- W2807515278 hasConceptScore W2807515278C119857082 @default.
- W2807515278 hasConceptScore W2807515278C138885662 @default.
- W2807515278 hasConceptScore W2807515278C153180895 @default.
- W2807515278 hasConceptScore W2807515278C154945302 @default.
- W2807515278 hasConceptScore W2807515278C189950617 @default.
- W2807515278 hasConceptScore W2807515278C2776401178 @default.
- W2807515278 hasConceptScore W2807515278C2777212361 @default.
- W2807515278 hasConceptScore W2807515278C36464697 @default.
- W2807515278 hasConceptScore W2807515278C41008148 @default.
- W2807515278 hasConceptScore W2807515278C41895202 @default.
- W2807515278 hasConceptScore W2807515278C81363708 @default.
- W2807515278 hasLocation W28075152781 @default.
- W2807515278 hasOpenAccess W2807515278 @default.
- W2807515278 hasPrimaryLocation W28075152781 @default.
- W2807515278 hasRelatedWork W2731899572 @default.
- W2807515278 hasRelatedWork W2999805992 @default.
- W2807515278 hasRelatedWork W3116150086 @default.
- W2807515278 hasRelatedWork W3133861977 @default.
- W2807515278 hasRelatedWork W4200173597 @default.
- W2807515278 hasRelatedWork W4223943233 @default.
- W2807515278 hasRelatedWork W4291897433 @default.
- W2807515278 hasRelatedWork W4312417841 @default.
- W2807515278 hasRelatedWork W4321369474 @default.
- W2807515278 hasRelatedWork W4380075502 @default.
- W2807515278 isParatext "false" @default.
- W2807515278 isRetracted "false" @default.
- W2807515278 magId "2807515278" @default.
- W2807515278 workType "article" @default.