Matches in SemOpenAlex for { <https://semopenalex.org/work/W4310554987> ?p ?o ?g. }
Showing items 1 to 75 of
75
with 100 items per page.
- W4310554987 abstract "Purpose: This study aims to explore training strategies to improve convolutional neural network-based image-to-image registration for abdominal imaging. Methods: Different training strategies, loss functions, and transfer learning schemes were considered. Furthermore, an augmentation layer which generates artificial training image pairs on-the-fly was proposed, in addition to a loss layer that enables dynamic loss weighting. Results: Guiding registration using segmentations in the training step proved beneficial for deep-learning-based image registration. Finetuning the pretrained model from the brain MRI dataset to the abdominal CT dataset further improved performance on the latter application, removing the need for a large dataset to yield satisfactory performance. Dynamic loss weighting also marginally improved performance, all without impacting inference runtime. Conclusion: Using simple concepts, we improved the performance of a commonly used deep image registration architecture, VoxelMorph. In future work, our framework, DDMR, should be validated on different datasets to further assess its value." @default.
- W4310554987 created "2022-12-12" @default.
- W4310554987 creator A5000199812 @default.
- W4310554987 creator A5010423987 @default.
- W4310554987 creator A5016840226 @default.
- W4310554987 creator A5036149753 @default.
- W4310554987 creator A5038367345 @default.
- W4310554987 creator A5040063476 @default.
- W4310554987 creator A5082365071 @default.
- W4310554987 creator A5090654106 @default.
- W4310554987 date "2022-11-28" @default.
- W4310554987 modified "2023-09-27" @default.
- W4310554987 title "Train smarter, not harder: learning deep abdominal CT registration on scarce data" @default.
- W4310554987 doi "https://doi.org/10.48550/arxiv.2211.15717" @default.
- W4310554987 hasPublicationYear "2022" @default.
- W4310554987 type Work @default.
- W4310554987 citedByCount "0" @default.
- W4310554987 crossrefType "posted-content" @default.
- W4310554987 hasAuthorship W4310554987A5000199812 @default.
- W4310554987 hasAuthorship W4310554987A5010423987 @default.
- W4310554987 hasAuthorship W4310554987A5016840226 @default.
- W4310554987 hasAuthorship W4310554987A5036149753 @default.
- W4310554987 hasAuthorship W4310554987A5038367345 @default.
- W4310554987 hasAuthorship W4310554987A5040063476 @default.
- W4310554987 hasAuthorship W4310554987A5082365071 @default.
- W4310554987 hasAuthorship W4310554987A5090654106 @default.
- W4310554987 hasBestOaLocation W43105549871 @default.
- W4310554987 hasConcept C108583219 @default.
- W4310554987 hasConcept C115961682 @default.
- W4310554987 hasConcept C119857082 @default.
- W4310554987 hasConcept C126838900 @default.
- W4310554987 hasConcept C150899416 @default.
- W4310554987 hasConcept C153180895 @default.
- W4310554987 hasConcept C154945302 @default.
- W4310554987 hasConcept C166704113 @default.
- W4310554987 hasConcept C183115368 @default.
- W4310554987 hasConcept C2776214188 @default.
- W4310554987 hasConcept C2984842247 @default.
- W4310554987 hasConcept C31972630 @default.
- W4310554987 hasConcept C41008148 @default.
- W4310554987 hasConcept C50644808 @default.
- W4310554987 hasConcept C71924100 @default.
- W4310554987 hasConcept C81363708 @default.
- W4310554987 hasConceptScore W4310554987C108583219 @default.
- W4310554987 hasConceptScore W4310554987C115961682 @default.
- W4310554987 hasConceptScore W4310554987C119857082 @default.
- W4310554987 hasConceptScore W4310554987C126838900 @default.
- W4310554987 hasConceptScore W4310554987C150899416 @default.
- W4310554987 hasConceptScore W4310554987C153180895 @default.
- W4310554987 hasConceptScore W4310554987C154945302 @default.
- W4310554987 hasConceptScore W4310554987C166704113 @default.
- W4310554987 hasConceptScore W4310554987C183115368 @default.
- W4310554987 hasConceptScore W4310554987C2776214188 @default.
- W4310554987 hasConceptScore W4310554987C2984842247 @default.
- W4310554987 hasConceptScore W4310554987C31972630 @default.
- W4310554987 hasConceptScore W4310554987C41008148 @default.
- W4310554987 hasConceptScore W4310554987C50644808 @default.
- W4310554987 hasConceptScore W4310554987C71924100 @default.
- W4310554987 hasConceptScore W4310554987C81363708 @default.
- W4310554987 hasLocation W43105549871 @default.
- W4310554987 hasOpenAccess W4310554987 @default.
- W4310554987 hasPrimaryLocation W43105549871 @default.
- W4310554987 hasRelatedWork W2738221750 @default.
- W4310554987 hasRelatedWork W2915754718 @default.
- W4310554987 hasRelatedWork W3016514588 @default.
- W4310554987 hasRelatedWork W3018421652 @default.
- W4310554987 hasRelatedWork W3021430260 @default.
- W4310554987 hasRelatedWork W3091976719 @default.
- W4310554987 hasRelatedWork W3166467183 @default.
- W4310554987 hasRelatedWork W3189091156 @default.
- W4310554987 hasRelatedWork W3192840557 @default.
- W4310554987 hasRelatedWork W4287009405 @default.
- W4310554987 isParatext "false" @default.
- W4310554987 isRetracted "false" @default.
- W4310554987 workType "article" @default.