Matches in SemOpenAlex for { <https://semopenalex.org/work/W3200339938> ?p ?o ?g. }
Showing items 1 to 91 of
91
with 100 items per page.
- W3200339938 abstract "Purpose. Brain Magnetic Resonance Images (MRIs) are essential for the diagnosis of neurological diseases. Recently, deep learning methods for unsupervised anomaly detection (UAD) have been proposed for the analysis of brain MRI. These methods rely on healthy brain MRIs and eliminate the requirement of pixel-wise annotated data compared to supervised deep learning. While a wide range of methods for UAD have been proposed, these methods are mostly 2D and only learn from MRI slices, disregarding that brain lesions are inherently 3D and the spatial context of MRI volumes remains unexploited. Methods. We investigate whether using increased spatial context by using MRI volumes combined with spatial erasing leads to improved unsupervised anomaly segmentation performance compared to learning from slices. We evaluate and compare 2D variational autoencoder (VAE) to their 3D counterpart, propose 3D input erasing, and systemically study the impact of the data set size on the performance. Results. Using two publicly available segmentation data sets for evaluation, 3D VAE outperform their 2D counterpart, highlighting the advantage of volumetric context. Also, our 3D erasing methods allow for further performance improvements. Our best performing 3D VAE with input erasing leads to an average DICE score of 31.40% compared to 25.76% for the 2D VAE. Conclusions. We propose 3D deep learning methods for UAD in brain MRI combined with 3D erasing and demonstrate that 3D methods clearly outperform their 2D counterpart for anomaly segmentation. Also, our spatial erasing method allows for further performance improvements and reduces the requirement for large data sets." @default.
- W3200339938 created "2021-09-27" @default.
- W3200339938 creator A5031373312 @default.
- W3200339938 creator A5057079579 @default.
- W3200339938 creator A5070915368 @default.
- W3200339938 creator A5083173883 @default.
- W3200339938 creator A5087348362 @default.
- W3200339938 date "2021-09-14" @default.
- W3200339938 modified "2023-09-27" @default.
- W3200339938 title "3-Dimensional Deep Learning with Spatial Erasing for Unsupervised Anomaly Segmentation in Brain MRI" @default.
- W3200339938 cites W1641498739 @default.
- W3200339938 cites W1959608418 @default.
- W3200339938 cites W2095705004 @default.
- W3200339938 cites W2147684460 @default.
- W3200339938 cites W2168199261 @default.
- W3200339938 cites W2621028221 @default.
- W3200339938 cites W2746314669 @default.
- W3200339938 cites W2751069891 @default.
- W3200339938 cites W2791423464 @default.
- W3200339938 cites W2791723176 @default.
- W3200339938 cites W2900298334 @default.
- W3200339938 cites W2914570111 @default.
- W3200339938 cites W2955425717 @default.
- W3200339938 cites W2963420272 @default.
- W3200339938 cites W2966606126 @default.
- W3200339938 cites W2997381454 @default.
- W3200339938 cites W3021584089 @default.
- W3200339938 cites W3028279406 @default.
- W3200339938 cites W3118868805 @default.
- W3200339938 cites W3196898581 @default.
- W3200339938 hasPublicationYear "2021" @default.
- W3200339938 type Work @default.
- W3200339938 sameAs 3200339938 @default.
- W3200339938 citedByCount "0" @default.
- W3200339938 crossrefType "posted-content" @default.
- W3200339938 hasAuthorship W3200339938A5031373312 @default.
- W3200339938 hasAuthorship W3200339938A5057079579 @default.
- W3200339938 hasAuthorship W3200339938A5070915368 @default.
- W3200339938 hasAuthorship W3200339938A5083173883 @default.
- W3200339938 hasAuthorship W3200339938A5087348362 @default.
- W3200339938 hasConcept C101738243 @default.
- W3200339938 hasConcept C108583219 @default.
- W3200339938 hasConcept C151730666 @default.
- W3200339938 hasConcept C153180895 @default.
- W3200339938 hasConcept C154945302 @default.
- W3200339938 hasConcept C2779343474 @default.
- W3200339938 hasConcept C41008148 @default.
- W3200339938 hasConcept C54170458 @default.
- W3200339938 hasConcept C64754055 @default.
- W3200339938 hasConcept C739882 @default.
- W3200339938 hasConcept C86803240 @default.
- W3200339938 hasConcept C89600930 @default.
- W3200339938 hasConceptScore W3200339938C101738243 @default.
- W3200339938 hasConceptScore W3200339938C108583219 @default.
- W3200339938 hasConceptScore W3200339938C151730666 @default.
- W3200339938 hasConceptScore W3200339938C153180895 @default.
- W3200339938 hasConceptScore W3200339938C154945302 @default.
- W3200339938 hasConceptScore W3200339938C2779343474 @default.
- W3200339938 hasConceptScore W3200339938C41008148 @default.
- W3200339938 hasConceptScore W3200339938C54170458 @default.
- W3200339938 hasConceptScore W3200339938C64754055 @default.
- W3200339938 hasConceptScore W3200339938C739882 @default.
- W3200339938 hasConceptScore W3200339938C86803240 @default.
- W3200339938 hasConceptScore W3200339938C89600930 @default.
- W3200339938 hasLocation W32003399381 @default.
- W3200339938 hasOpenAccess W3200339938 @default.
- W3200339938 hasPrimaryLocation W32003399381 @default.
- W3200339938 hasRelatedWork W1509389319 @default.
- W3200339938 hasRelatedWork W2027839406 @default.
- W3200339938 hasRelatedWork W2518214538 @default.
- W3200339938 hasRelatedWork W2729876886 @default.
- W3200339938 hasRelatedWork W2760516000 @default.
- W3200339938 hasRelatedWork W2781600508 @default.
- W3200339938 hasRelatedWork W2787015083 @default.
- W3200339938 hasRelatedWork W2796762894 @default.
- W3200339938 hasRelatedWork W2944052559 @default.
- W3200339938 hasRelatedWork W2979699329 @default.
- W3200339938 hasRelatedWork W3013085919 @default.
- W3200339938 hasRelatedWork W3026376770 @default.
- W3200339938 hasRelatedWork W3111577664 @default.
- W3200339938 hasRelatedWork W3122598341 @default.
- W3200339938 hasRelatedWork W3138697998 @default.
- W3200339938 hasRelatedWork W3157690401 @default.
- W3200339938 hasRelatedWork W3158191311 @default.
- W3200339938 hasRelatedWork W3178092667 @default.
- W3200339938 hasRelatedWork W3198811856 @default.
- W3200339938 hasRelatedWork W3200457825 @default.
- W3200339938 isParatext "false" @default.
- W3200339938 isRetracted "false" @default.
- W3200339938 magId "3200339938" @default.
- W3200339938 workType "article" @default.