Matches in SemOpenAlex for { <https://semopenalex.org/work/W3203363563> ?p ?o ?g. }
Showing items 1 to 76 of
76
with 100 items per page.
- W3203363563 endingPage "10" @default.
- W3203363563 startingPage "1" @default.
- W3203363563 abstract "Uncertainty assessment has gained rapid interest in medical image analysis. A popular technique to compute epistemic uncertainty is the Monte-Carlo (MC) dropout technique. From a network with MC dropout and a single input, multiple output can be sampled. Various methods can be used to obtain epistemic uncertainty maps from those multiple outputs. In the case, of multi-class segmentation the number of methods is even larger as epistemic uncertainty can be computed voxelwise per class or voxelwise per image. This paper highlights a systematic approach to define and quantitatively compare those methods in those two different contexts: class specific epistemic uncertainty maps (one value per image, voxel and class) and combined epistemic uncertainty maps (one value per image and voxel). We applied this quantitative analysis to a multi-class segmentation of the carotid artery lumen and vessel wall, on a multi-center, multi-scanner, multi-sequence dataset of Magnetic Resonance (MR) images. We validated our analysis over 144 sets of hyperparameters of a model. Our main analysis consider the relationship between the order of the voxels of an epistemic uncertainty map and the misclassification of the prediction. Under this consideration, the comparison of combined uncertainty maps reveals that the multi-class entropy and the multi-class mutual information statistically out-perform the other combined uncertainty maps under study (the averaged entropy, the averaged variance, the similarity Bhattacharya coefficient and the similarity Kullback-Leibler divergence). In a class specific scenario, the one versus all entropy statistically out-performs the class-wise entropy, the class-wise variance and the one versus all mutual information. The class-wise entropy statistically out-performs the other class specific uncertainty maps in term of calibration. We made a python package available at to reproduce our analysis on different data and tasks." @default.
- W3203363563 created "2021-10-11" @default.
- W3203363563 creator A5006225939 @default.
- W3203363563 creator A5009602483 @default.
- W3203363563 creator A5022417082 @default.
- W3203363563 creator A5023677285 @default.
- W3203363563 creator A5051576018 @default.
- W3203363563 creator A5059145563 @default.
- W3203363563 creator A5080311936 @default.
- W3203363563 date "2021-09-28" @default.
- W3203363563 modified "2023-09-23" @default.
- W3203363563 title "A Quantitative Comparison of Epistemic Uncertainty Maps Applied to Multi-Class Segmentation" @default.
- W3203363563 hasPublicationYear "2021" @default.
- W3203363563 type Work @default.
- W3203363563 sameAs 3203363563 @default.
- W3203363563 citedByCount "0" @default.
- W3203363563 crossrefType "journal-article" @default.
- W3203363563 hasAuthorship W3203363563A5006225939 @default.
- W3203363563 hasAuthorship W3203363563A5009602483 @default.
- W3203363563 hasAuthorship W3203363563A5022417082 @default.
- W3203363563 hasAuthorship W3203363563A5023677285 @default.
- W3203363563 hasAuthorship W3203363563A5051576018 @default.
- W3203363563 hasAuthorship W3203363563A5059145563 @default.
- W3203363563 hasAuthorship W3203363563A5080311936 @default.
- W3203363563 hasConcept C105795698 @default.
- W3203363563 hasConcept C106301342 @default.
- W3203363563 hasConcept C121332964 @default.
- W3203363563 hasConcept C153180895 @default.
- W3203363563 hasConcept C154945302 @default.
- W3203363563 hasConcept C184898388 @default.
- W3203363563 hasConcept C32230216 @default.
- W3203363563 hasConcept C33923547 @default.
- W3203363563 hasConcept C41008148 @default.
- W3203363563 hasConcept C54170458 @default.
- W3203363563 hasConcept C62520636 @default.
- W3203363563 hasConcept C89600930 @default.
- W3203363563 hasConceptScore W3203363563C105795698 @default.
- W3203363563 hasConceptScore W3203363563C106301342 @default.
- W3203363563 hasConceptScore W3203363563C121332964 @default.
- W3203363563 hasConceptScore W3203363563C153180895 @default.
- W3203363563 hasConceptScore W3203363563C154945302 @default.
- W3203363563 hasConceptScore W3203363563C184898388 @default.
- W3203363563 hasConceptScore W3203363563C32230216 @default.
- W3203363563 hasConceptScore W3203363563C33923547 @default.
- W3203363563 hasConceptScore W3203363563C41008148 @default.
- W3203363563 hasConceptScore W3203363563C54170458 @default.
- W3203363563 hasConceptScore W3203363563C62520636 @default.
- W3203363563 hasConceptScore W3203363563C89600930 @default.
- W3203363563 hasOpenAccess W3203363563 @default.
- W3203363563 hasRelatedWork W1479893190 @default.
- W3203363563 hasRelatedWork W1985335554 @default.
- W3203363563 hasRelatedWork W2051875912 @default.
- W3203363563 hasRelatedWork W2066805929 @default.
- W3203363563 hasRelatedWork W2100030469 @default.
- W3203363563 hasRelatedWork W2107565029 @default.
- W3203363563 hasRelatedWork W2115823281 @default.
- W3203363563 hasRelatedWork W2240972438 @default.
- W3203363563 hasRelatedWork W2429473218 @default.
- W3203363563 hasRelatedWork W2611832709 @default.
- W3203363563 hasRelatedWork W2758801564 @default.
- W3203363563 hasRelatedWork W2903349397 @default.
- W3203363563 hasRelatedWork W2904657988 @default.
- W3203363563 hasRelatedWork W2980008337 @default.
- W3203363563 hasRelatedWork W2995789551 @default.
- W3203363563 hasRelatedWork W3002954653 @default.
- W3203363563 hasRelatedWork W3025924906 @default.
- W3203363563 hasRelatedWork W3087201208 @default.
- W3203363563 hasRelatedWork W3132744937 @default.
- W3203363563 hasRelatedWork W3194042225 @default.
- W3203363563 hasVolume "013" @default.
- W3203363563 isParatext "false" @default.
- W3203363563 isRetracted "false" @default.
- W3203363563 magId "3203363563" @default.
- W3203363563 workType "article" @default.