Matches in SemOpenAlex for { <https://semopenalex.org/work/W3037451087> ?p ?o ?g. }
- W3037451087 abstract "The advanced magnetic resonance (MR) image reconstructions such as the compressed sensing and subspace-based imaging are considered as large-scale, iterative, optimization problems. Given the large number of reconstructions required by the practical clinical usage, the computation time of these advanced reconstruction methods is often unacceptable. In this work, we propose using Google's Tensor Processing Units (TPUs) to accelerate the MR image reconstruction. TPU is an application-specific integrated circuit (ASIC) for machine learning applications, which has recently been used to solve large-scale scientific computing problems. As proof-of-concept, we implement the alternating direction method of multipliers (ADMM) in TensorFlow to reconstruct images on TPUs. The reconstruction is based on multi-channel, sparsely sampled, and radial-trajectory $k$-space data with sparsity constraints. The forward and inverse non-uniform Fourier transform operations are formulated in terms of matrix multiplications as in the discrete Fourier transform. The sparsifying transform and its adjoint operations are formulated as convolutions. The data decomposition is applied to the measured $k$-space data such that the aforementioned tensor operations are localized within individual TPU cores. The data decomposition and the inter-core communication strategy are designed in accordance with the TPU interconnect network topology in order to minimize the communication time. The accuracy and the high parallel efficiency of the proposed TPU-based image reconstruction method are demonstrated through numerical examples." @default.
- W3037451087 created "2020-07-02" @default.
- W3037451087 creator A5044388284 @default.
- W3037451087 creator A5054510529 @default.
- W3037451087 creator A5062139423 @default.
- W3037451087 creator A5062325991 @default.
- W3037451087 creator A5069758006 @default.
- W3037451087 date "2020-06-24" @default.
- W3037451087 modified "2023-09-23" @default.
- W3037451087 title "Accelerating MRI Reconstruction on TPUs" @default.
- W3037451087 cites W1964397874 @default.
- W3037451087 cites W1993917215 @default.
- W3037451087 cites W2016641965 @default.
- W3037451087 cites W2036343507 @default.
- W3037451087 cites W2055844022 @default.
- W3037451087 cites W2080236984 @default.
- W3037451087 cites W2099944953 @default.
- W3037451087 cites W2101507236 @default.
- W3037451087 cites W2101675075 @default.
- W3037451087 cites W2115723093 @default.
- W3037451087 cites W2117406892 @default.
- W3037451087 cites W2124021916 @default.
- W3037451087 cites W2129638195 @default.
- W3037451087 cites W2142808775 @default.
- W3037451087 cites W2145096794 @default.
- W3037451087 cites W2150903265 @default.
- W3037451087 cites W2157931750 @default.
- W3037451087 cites W2164278908 @default.
- W3037451087 cites W2170255449 @default.
- W3037451087 cites W2296616510 @default.
- W3037451087 cites W2399267115 @default.
- W3037451087 cites W2525778437 @default.
- W3037451087 cites W2588773459 @default.
- W3037451087 cites W2606722458 @default.
- W3037451087 cites W2620440553 @default.
- W3037451087 cites W2752792052 @default.
- W3037451087 cites W2777638777 @default.
- W3037451087 cites W2790485410 @default.
- W3037451087 cites W2948528583 @default.
- W3037451087 cites W2953212265 @default.
- W3037451087 cites W2983490771 @default.
- W3037451087 cites W3004740094 @default.
- W3037451087 cites W2167233877 @default.
- W3037451087 cites W2168887049 @default.
- W3037451087 hasPublicationYear "2020" @default.
- W3037451087 type Work @default.
- W3037451087 sameAs 3037451087 @default.
- W3037451087 citedByCount "0" @default.
- W3037451087 crossrefType "posted-content" @default.
- W3037451087 hasAuthorship W3037451087A5044388284 @default.
- W3037451087 hasAuthorship W3037451087A5054510529 @default.
- W3037451087 hasAuthorship W3037451087A5062139423 @default.
- W3037451087 hasAuthorship W3037451087A5062325991 @default.
- W3037451087 hasAuthorship W3037451087A5069758006 @default.
- W3037451087 hasConcept C102519508 @default.
- W3037451087 hasConcept C11413529 @default.
- W3037451087 hasConcept C126255220 @default.
- W3037451087 hasConcept C134306372 @default.
- W3037451087 hasConcept C141379421 @default.
- W3037451087 hasConcept C155281189 @default.
- W3037451087 hasConcept C173608175 @default.
- W3037451087 hasConcept C200127275 @default.
- W3037451087 hasConcept C202444582 @default.
- W3037451087 hasConcept C203249530 @default.
- W3037451087 hasConcept C2524010 @default.
- W3037451087 hasConcept C31972630 @default.
- W3037451087 hasConcept C33923547 @default.
- W3037451087 hasConcept C41008148 @default.
- W3037451087 hasConcept C459310 @default.
- W3037451087 hasConcept C80444323 @default.
- W3037451087 hasConceptScore W3037451087C102519508 @default.
- W3037451087 hasConceptScore W3037451087C11413529 @default.
- W3037451087 hasConceptScore W3037451087C126255220 @default.
- W3037451087 hasConceptScore W3037451087C134306372 @default.
- W3037451087 hasConceptScore W3037451087C141379421 @default.
- W3037451087 hasConceptScore W3037451087C155281189 @default.
- W3037451087 hasConceptScore W3037451087C173608175 @default.
- W3037451087 hasConceptScore W3037451087C200127275 @default.
- W3037451087 hasConceptScore W3037451087C202444582 @default.
- W3037451087 hasConceptScore W3037451087C203249530 @default.
- W3037451087 hasConceptScore W3037451087C2524010 @default.
- W3037451087 hasConceptScore W3037451087C31972630 @default.
- W3037451087 hasConceptScore W3037451087C33923547 @default.
- W3037451087 hasConceptScore W3037451087C41008148 @default.
- W3037451087 hasConceptScore W3037451087C459310 @default.
- W3037451087 hasConceptScore W3037451087C80444323 @default.
- W3037451087 hasLocation W30374510871 @default.
- W3037451087 hasOpenAccess W3037451087 @default.
- W3037451087 hasPrimaryLocation W30374510871 @default.
- W3037451087 hasRelatedWork W1588180435 @default.
- W3037451087 hasRelatedWork W1988504501 @default.
- W3037451087 hasRelatedWork W1992112669 @default.
- W3037451087 hasRelatedWork W2033336457 @default.
- W3037451087 hasRelatedWork W2102909376 @default.
- W3037451087 hasRelatedWork W2126407179 @default.
- W3037451087 hasRelatedWork W2317923507 @default.
- W3037451087 hasRelatedWork W2339161400 @default.
- W3037451087 hasRelatedWork W2507398220 @default.
- W3037451087 hasRelatedWork W2527299897 @default.
- W3037451087 hasRelatedWork W2740906587 @default.