Matches in SemOpenAlex for { <https://semopenalex.org/work/W4384388077> ?p ?o ?g. }
Showing items 1 to 69 of
69
with 100 items per page.
- W4384388077 abstract "Meta-learning has recently been an emerging data-efficient learning technique for various medical imaging operations and has helped advance contemporary deep learning models. Furthermore, meta-learning enhances the knowledge generalization of the imaging tasks by learning both shared and discriminative weights for various configurations of imaging tasks. However, existing meta-learning models attempt to learn a single set of weight initializations of a neural network that might be restrictive for multimodal data. This work aims to develop a multimodal meta-learning model for image reconstruction, which augments meta-learning with evolutionary capabilities to encompass diverse acquisition settings of multimodal data. Our proposed model called KM-MAML (Kernel Modulation-based Multimodal Meta-Learning), has hypernetworks that evolve to generate mode-specific weights. These weights provide the mode-specific inductive bias for multiple modes by re-calibrating each kernel of the base network for image reconstruction via a low-rank kernel modulation operation. We incorporate gradient-based meta-learning (GBML) in the contextual space to update the weights of the hypernetworks for different modes. The hypernetworks and the reconstruction network in the GBML setting provide discriminative mode-specific features and low-level image features, respectively. Experiments on multi-contrast MRI reconstruction show that our model, (i) exhibits superior reconstruction performance over joint training, other meta-learning methods, and context-specific MRI reconstruction methods, and (ii) better adaptation capabilities with improvement margins of 0.5 dB in PSNR and 0.01 in SSIM. Besides, a representation analysis with U-Net shows that kernel modulation infuses 80% of mode-specific representation changes in the high-resolution layers. Our source code is available at https://github.com/sriprabhar/KM-MAML/." @default.
- W4384388077 created "2023-07-15" @default.
- W4384388077 creator A5001368638 @default.
- W4384388077 creator A5032454627 @default.
- W4384388077 creator A5062250937 @default.
- W4384388077 creator A5083037666 @default.
- W4384388077 date "2023-07-13" @default.
- W4384388077 modified "2023-09-27" @default.
- W4384388077 title "Generalizing Supervised Deep Learning MRI Reconstruction to Multiple and Unseen Contrasts using Meta-Learning Hypernetworks" @default.
- W4384388077 doi "https://doi.org/10.48550/arxiv.2307.06771" @default.
- W4384388077 hasPublicationYear "2023" @default.
- W4384388077 type Work @default.
- W4384388077 citedByCount "0" @default.
- W4384388077 crossrefType "posted-content" @default.
- W4384388077 hasAuthorship W4384388077A5001368638 @default.
- W4384388077 hasAuthorship W4384388077A5032454627 @default.
- W4384388077 hasAuthorship W4384388077A5062250937 @default.
- W4384388077 hasAuthorship W4384388077A5083037666 @default.
- W4384388077 hasBestOaLocation W43843880771 @default.
- W4384388077 hasConcept C108583219 @default.
- W4384388077 hasConcept C114614502 @default.
- W4384388077 hasConcept C119857082 @default.
- W4384388077 hasConcept C134306372 @default.
- W4384388077 hasConcept C153180895 @default.
- W4384388077 hasConcept C154945302 @default.
- W4384388077 hasConcept C162324750 @default.
- W4384388077 hasConcept C177148314 @default.
- W4384388077 hasConcept C187736073 @default.
- W4384388077 hasConcept C2780451532 @default.
- W4384388077 hasConcept C2781002164 @default.
- W4384388077 hasConcept C33923547 @default.
- W4384388077 hasConcept C41008148 @default.
- W4384388077 hasConcept C50644808 @default.
- W4384388077 hasConcept C59404180 @default.
- W4384388077 hasConcept C74193536 @default.
- W4384388077 hasConcept C97931131 @default.
- W4384388077 hasConceptScore W4384388077C108583219 @default.
- W4384388077 hasConceptScore W4384388077C114614502 @default.
- W4384388077 hasConceptScore W4384388077C119857082 @default.
- W4384388077 hasConceptScore W4384388077C134306372 @default.
- W4384388077 hasConceptScore W4384388077C153180895 @default.
- W4384388077 hasConceptScore W4384388077C154945302 @default.
- W4384388077 hasConceptScore W4384388077C162324750 @default.
- W4384388077 hasConceptScore W4384388077C177148314 @default.
- W4384388077 hasConceptScore W4384388077C187736073 @default.
- W4384388077 hasConceptScore W4384388077C2780451532 @default.
- W4384388077 hasConceptScore W4384388077C2781002164 @default.
- W4384388077 hasConceptScore W4384388077C33923547 @default.
- W4384388077 hasConceptScore W4384388077C41008148 @default.
- W4384388077 hasConceptScore W4384388077C50644808 @default.
- W4384388077 hasConceptScore W4384388077C59404180 @default.
- W4384388077 hasConceptScore W4384388077C74193536 @default.
- W4384388077 hasConceptScore W4384388077C97931131 @default.
- W4384388077 hasLocation W43843880771 @default.
- W4384388077 hasOpenAccess W4384388077 @default.
- W4384388077 hasPrimaryLocation W43843880771 @default.
- W4384388077 hasRelatedWork W130490334 @default.
- W4384388077 hasRelatedWork W1914651075 @default.
- W4384388077 hasRelatedWork W2061273563 @default.
- W4384388077 hasRelatedWork W2353457699 @default.
- W4384388077 hasRelatedWork W2507989420 @default.
- W4384388077 hasRelatedWork W2773120646 @default.
- W4384388077 hasRelatedWork W2905846897 @default.
- W4384388077 hasRelatedWork W2998168123 @default.
- W4384388077 hasRelatedWork W3025754292 @default.
- W4384388077 hasRelatedWork W4287995534 @default.
- W4384388077 isParatext "false" @default.
- W4384388077 isRetracted "false" @default.
- W4384388077 workType "article" @default.