Matches in SemOpenAlex for { <https://semopenalex.org/work/W3188455524> ?p ?o ?g. }
Showing items 1 to 74 of
74
with 100 items per page.
- W3188455524 endingPage "733" @default.
- W3188455524 startingPage "724" @default.
- W3188455524 abstract "Convolutional Neural Networks (CNNs) are highly effective for image reconstruction problems. Typically, CNNs are trained on large amounts of training images. Recently, however, un-trained CNNs such as the Deep Image Prior and Deep Decoder have achieved excellent performance for image reconstruction problems such as denoising and inpainting, without using any training data —except leveraging a few samples for hyper-parameter tuning. Motivated by this development, we address the reconstruction problem arising in accelerated MRI with un-trained neural networks. We propose a highly optimized un-trained recovery approach based on a variation of the Deep Decoder and show that it significantly outperforms other un-trained methods, in particular sparsity-based classical compressed sensing methods and naive applications of un-trained neural networks in terms of reconstruction performance. We also compare performance (both in terms of reconstruction accuracy and computational cost) in an ideal setup for trained methods, specifically on the fastMRI dataset, where the training and test data come from the same distribution. Here, we find that our un-trained algorithm achieves similar performance to a baseline trained neural network, but a state-of-the-art trained network outperforms the un-trained one. Finally, we perform a comparison on a non-ideal setup where the train and test distributions are slightly different, and find that our un-trained method achieves similar performance to a state-of-the-art accelerated MRI reconstruction method." @default.
- W3188455524 created "2021-08-16" @default.
- W3188455524 creator A5003606899 @default.
- W3188455524 creator A5077284648 @default.
- W3188455524 date "2021-01-01" @default.
- W3188455524 modified "2023-10-02" @default.
- W3188455524 title "Accelerated MRI With Un-Trained Neural Networks" @default.
- W3188455524 cites W1497904071 @default.
- W3188455524 cites W1580389772 @default.
- W3188455524 cites W1901129140 @default.
- W3188455524 cites W2029816571 @default.
- W3188455524 cites W2037642501 @default.
- W3188455524 cites W2046119925 @default.
- W3188455524 cites W2117649283 @default.
- W3188455524 cites W2133665775 @default.
- W3188455524 cites W2145020729 @default.
- W3188455524 cites W2165142794 @default.
- W3188455524 cites W2916033043 @default.
- W3188455524 cites W2950936580 @default.
- W3188455524 cites W2962903101 @default.
- W3188455524 cites W2998280611 @default.
- W3188455524 cites W2999511788 @default.
- W3188455524 cites W2999819348 @default.
- W3188455524 cites W3017940382 @default.
- W3188455524 cites W3020963216 @default.
- W3188455524 cites W3034223847 @default.
- W3188455524 cites W3039236647 @default.
- W3188455524 cites W3092530614 @default.
- W3188455524 cites W3101335681 @default.
- W3188455524 cites W3157723514 @default.
- W3188455524 cites W3163954017 @default.
- W3188455524 doi "https://doi.org/10.1109/tci.2021.3097596" @default.
- W3188455524 hasPublicationYear "2021" @default.
- W3188455524 type Work @default.
- W3188455524 sameAs 3188455524 @default.
- W3188455524 citedByCount "35" @default.
- W3188455524 countsByYear W31884555242021 @default.
- W3188455524 countsByYear W31884555242022 @default.
- W3188455524 countsByYear W31884555242023 @default.
- W3188455524 crossrefType "journal-article" @default.
- W3188455524 hasAuthorship W3188455524A5003606899 @default.
- W3188455524 hasAuthorship W3188455524A5077284648 @default.
- W3188455524 hasBestOaLocation W31884555241 @default.
- W3188455524 hasConcept C153180895 @default.
- W3188455524 hasConcept C154945302 @default.
- W3188455524 hasConcept C31972630 @default.
- W3188455524 hasConcept C41008148 @default.
- W3188455524 hasConcept C50644808 @default.
- W3188455524 hasConceptScore W3188455524C153180895 @default.
- W3188455524 hasConceptScore W3188455524C154945302 @default.
- W3188455524 hasConceptScore W3188455524C31972630 @default.
- W3188455524 hasConceptScore W3188455524C41008148 @default.
- W3188455524 hasConceptScore W3188455524C50644808 @default.
- W3188455524 hasLocation W31884555241 @default.
- W3188455524 hasLocation W31884555242 @default.
- W3188455524 hasOpenAccess W3188455524 @default.
- W3188455524 hasPrimaryLocation W31884555241 @default.
- W3188455524 hasRelatedWork W1891287906 @default.
- W3188455524 hasRelatedWork W1969923398 @default.
- W3188455524 hasRelatedWork W2036807459 @default.
- W3188455524 hasRelatedWork W2058170566 @default.
- W3188455524 hasRelatedWork W2170022336 @default.
- W3188455524 hasRelatedWork W2229312674 @default.
- W3188455524 hasRelatedWork W258625772 @default.
- W3188455524 hasRelatedWork W2755342338 @default.
- W3188455524 hasRelatedWork W2772917594 @default.
- W3188455524 hasRelatedWork W3116076068 @default.
- W3188455524 hasVolume "7" @default.
- W3188455524 isParatext "false" @default.
- W3188455524 isRetracted "false" @default.
- W3188455524 magId "3188455524" @default.
- W3188455524 workType "article" @default.