Matches in SemOpenAlex for { <https://semopenalex.org/work/W3093568609> ?p ?o ?g. }
Showing items 1 to 93 of
93
with 100 items per page.
- W3093568609 endingPage "734" @default.
- W3093568609 startingPage "701" @default.
- W3093568609 abstract "Abstract Deep learning (DL), which involves powerful black box predictors, has achieved a remarkable performance in medical image analysis, such as segmentation and classification for diagnosis. However, in spite of these successes, these methods focus exclusively on improving the accuracy of point predictions without assessing the quality of their outputs. Knowing how much confidence there is in a prediction is essential for gaining clinicians' trust in the technology. In this article, we propose an uncertainty estimation framework, called MC‐DropWeights, to approximate Bayesian inference in DL by imposing a Bernoulli distribution on the incoming or outgoing weights of the model, including neurones. We demonstrate that by decomposing predictive probabilities into two main types of uncertainty, aleatoric and epistemic, using the Bayesian Residual U‐Net (BRUNet) in image segmentation. Approximation methods in Bayesian DL suffer from the “mode collapse” phenomenon in variational inference. To address this problem, we propose a model which Ensembles of Monte‐Carlo DropWeights by varying the DropWeights rate. In segmentation, we introduce a predictive uncertainty estimator, which takes the mean of the standard deviations of the class probabilities associated with every class. However, in classification, we need an alternative approach since the predictive probabilities from a forward pass through the model does not capture uncertainty. The entropy of the predictive distribution is a measure of uncertainty, but its exponential depends on sample size. The plug‐in estimate in mutual information is subject to sampling bias. We propose Jackknife resampling, to correct for sample bias, which improves estimating uncertainty quality in image classification. We demonstrate that our deep ensemble MC‐DropWeights method, using the bias‐corrected estimator produces an equally good or better result in both quantified uncertainty estimation and quality of uncertainty estimates than approximate Bayesian neural networks in practice." @default.
- W3093568609 created "2020-10-29" @default.
- W3093568609 creator A5004397914 @default.
- W3093568609 creator A5013404589 @default.
- W3093568609 creator A5050505627 @default.
- W3093568609 creator A5089973536 @default.
- W3093568609 date "2020-10-22" @default.
- W3093568609 modified "2023-10-16" @default.
- W3093568609 title "Estimating uncertainty in deep learning for reporting confidence to clinicians in medical image segmentation and diseases detection" @default.
- W3093568609 cites W1516339661 @default.
- W3093568609 cites W1534477342 @default.
- W3093568609 cites W1677182931 @default.
- W3093568609 cites W1901129140 @default.
- W3093568609 cites W1970120446 @default.
- W3093568609 cites W1995875735 @default.
- W3093568609 cites W2047229728 @default.
- W3093568609 cites W2051765910 @default.
- W3093568609 cites W2079473986 @default.
- W3093568609 cites W2083927153 @default.
- W3093568609 cites W2111051539 @default.
- W3093568609 cites W2115599658 @default.
- W3093568609 cites W2123957845 @default.
- W3093568609 cites W2145889472 @default.
- W3093568609 cites W2146705998 @default.
- W3093568609 cites W2148743296 @default.
- W3093568609 cites W2163586890 @default.
- W3093568609 cites W2548342201 @default.
- W3093568609 cites W2592929672 @default.
- W3093568609 cites W2790913410 @default.
- W3093568609 cites W2919115771 @default.
- W3093568609 cites W2962949934 @default.
- W3093568609 cites W2963881378 @default.
- W3093568609 cites W2964706145 @default.
- W3093568609 cites W2965563166 @default.
- W3093568609 cites W4206273459 @default.
- W3093568609 cites W4239315130 @default.
- W3093568609 doi "https://doi.org/10.1111/coin.12411" @default.
- W3093568609 hasPublicationYear "2020" @default.
- W3093568609 type Work @default.
- W3093568609 sameAs 3093568609 @default.
- W3093568609 citedByCount "26" @default.
- W3093568609 countsByYear W30935686092020 @default.
- W3093568609 countsByYear W30935686092021 @default.
- W3093568609 countsByYear W30935686092022 @default.
- W3093568609 countsByYear W30935686092023 @default.
- W3093568609 crossrefType "journal-article" @default.
- W3093568609 hasAuthorship W3093568609A5004397914 @default.
- W3093568609 hasAuthorship W3093568609A5013404589 @default.
- W3093568609 hasAuthorship W3093568609A5050505627 @default.
- W3093568609 hasAuthorship W3093568609A5089973536 @default.
- W3093568609 hasBestOaLocation W30935686091 @default.
- W3093568609 hasConcept C105795698 @default.
- W3093568609 hasConcept C107673813 @default.
- W3093568609 hasConcept C119857082 @default.
- W3093568609 hasConcept C124504099 @default.
- W3093568609 hasConcept C153180895 @default.
- W3093568609 hasConcept C154945302 @default.
- W3093568609 hasConcept C177769412 @default.
- W3093568609 hasConcept C33923547 @default.
- W3093568609 hasConcept C41008148 @default.
- W3093568609 hasConcept C89600930 @default.
- W3093568609 hasConceptScore W3093568609C105795698 @default.
- W3093568609 hasConceptScore W3093568609C107673813 @default.
- W3093568609 hasConceptScore W3093568609C119857082 @default.
- W3093568609 hasConceptScore W3093568609C124504099 @default.
- W3093568609 hasConceptScore W3093568609C153180895 @default.
- W3093568609 hasConceptScore W3093568609C154945302 @default.
- W3093568609 hasConceptScore W3093568609C177769412 @default.
- W3093568609 hasConceptScore W3093568609C33923547 @default.
- W3093568609 hasConceptScore W3093568609C41008148 @default.
- W3093568609 hasConceptScore W3093568609C89600930 @default.
- W3093568609 hasIssue "2" @default.
- W3093568609 hasLocation W30935686091 @default.
- W3093568609 hasLocation W30935686092 @default.
- W3093568609 hasOpenAccess W3093568609 @default.
- W3093568609 hasPrimaryLocation W30935686091 @default.
- W3093568609 hasRelatedWork W1718066205 @default.
- W3093568609 hasRelatedWork W2015518264 @default.
- W3093568609 hasRelatedWork W2135187896 @default.
- W3093568609 hasRelatedWork W2147201983 @default.
- W3093568609 hasRelatedWork W2160108762 @default.
- W3093568609 hasRelatedWork W2168271748 @default.
- W3093568609 hasRelatedWork W2562263695 @default.
- W3093568609 hasRelatedWork W3160546271 @default.
- W3093568609 hasRelatedWork W4311646318 @default.
- W3093568609 hasRelatedWork W2795035211 @default.
- W3093568609 hasVolume "37" @default.
- W3093568609 isParatext "false" @default.
- W3093568609 isRetracted "false" @default.
- W3093568609 magId "3093568609" @default.
- W3093568609 workType "article" @default.