Cargando…

High-Performance 3D Compressive Sensing MRI Reconstruction Using Many-Core Architectures

Compressive sensing (CS) describes how sparse signals can be accurately reconstructed from many fewer samples than required by the Nyquist criterion. Since MRI scan duration is proportional to the number of acquired samples, CS has been gaining significant attention in MRI. However, the computationa...

Descripción completa

Detalles Bibliográficos
Autores principales: Kim, Daehyun, Trzasko, Joshua, Smelyanskiy, Mikhail, Haider, Clifton, Dubey, Pradeep, Manduca, Armando
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Hindawi Publishing Corporation 2011
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3172979/
https://www.ncbi.nlm.nih.gov/pubmed/21922017
http://dx.doi.org/10.1155/2011/473128
_version_ 1782211922833178624
author Kim, Daehyun
Trzasko, Joshua
Smelyanskiy, Mikhail
Haider, Clifton
Dubey, Pradeep
Manduca, Armando
author_facet Kim, Daehyun
Trzasko, Joshua
Smelyanskiy, Mikhail
Haider, Clifton
Dubey, Pradeep
Manduca, Armando
author_sort Kim, Daehyun
collection PubMed
description Compressive sensing (CS) describes how sparse signals can be accurately reconstructed from many fewer samples than required by the Nyquist criterion. Since MRI scan duration is proportional to the number of acquired samples, CS has been gaining significant attention in MRI. However, the computationally intensive nature of CS reconstructions has precluded their use in routine clinical practice. In this work, we investigate how different throughput-oriented architectures can benefit one CS algorithm and what levels of acceleration are feasible on different modern platforms. We demonstrate that a CUDA-based code running on an NVIDIA Tesla C2050 GPU can reconstruct a 256 × 160 × 80 volume from an 8-channel acquisition in 19 seconds, which is in itself a significant improvement over the state of the art. We then show that Intel's Knights Ferry can perform the same 3D MRI reconstruction in only 12 seconds, bringing CS methods even closer to clinical viability.
format Online
Article
Text
id pubmed-3172979
institution National Center for Biotechnology Information
language English
publishDate 2011
publisher Hindawi Publishing Corporation
record_format MEDLINE/PubMed
spelling pubmed-31729792011-09-15 High-Performance 3D Compressive Sensing MRI Reconstruction Using Many-Core Architectures Kim, Daehyun Trzasko, Joshua Smelyanskiy, Mikhail Haider, Clifton Dubey, Pradeep Manduca, Armando Int J Biomed Imaging Research Article Compressive sensing (CS) describes how sparse signals can be accurately reconstructed from many fewer samples than required by the Nyquist criterion. Since MRI scan duration is proportional to the number of acquired samples, CS has been gaining significant attention in MRI. However, the computationally intensive nature of CS reconstructions has precluded their use in routine clinical practice. In this work, we investigate how different throughput-oriented architectures can benefit one CS algorithm and what levels of acceleration are feasible on different modern platforms. We demonstrate that a CUDA-based code running on an NVIDIA Tesla C2050 GPU can reconstruct a 256 × 160 × 80 volume from an 8-channel acquisition in 19 seconds, which is in itself a significant improvement over the state of the art. We then show that Intel's Knights Ferry can perform the same 3D MRI reconstruction in only 12 seconds, bringing CS methods even closer to clinical viability. Hindawi Publishing Corporation 2011 2011-09-14 /pmc/articles/PMC3172979/ /pubmed/21922017 http://dx.doi.org/10.1155/2011/473128 Text en Copyright © 2011 Daehyun Kim et al. https://creativecommons.org/licenses/by/3.0/ This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
spellingShingle Research Article
Kim, Daehyun
Trzasko, Joshua
Smelyanskiy, Mikhail
Haider, Clifton
Dubey, Pradeep
Manduca, Armando
High-Performance 3D Compressive Sensing MRI Reconstruction Using Many-Core Architectures
title High-Performance 3D Compressive Sensing MRI Reconstruction Using Many-Core Architectures
title_full High-Performance 3D Compressive Sensing MRI Reconstruction Using Many-Core Architectures
title_fullStr High-Performance 3D Compressive Sensing MRI Reconstruction Using Many-Core Architectures
title_full_unstemmed High-Performance 3D Compressive Sensing MRI Reconstruction Using Many-Core Architectures
title_short High-Performance 3D Compressive Sensing MRI Reconstruction Using Many-Core Architectures
title_sort high-performance 3d compressive sensing mri reconstruction using many-core architectures
topic Research Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3172979/
https://www.ncbi.nlm.nih.gov/pubmed/21922017
http://dx.doi.org/10.1155/2011/473128
work_keys_str_mv AT kimdaehyun highperformance3dcompressivesensingmrireconstructionusingmanycorearchitectures
AT trzaskojoshua highperformance3dcompressivesensingmrireconstructionusingmanycorearchitectures
AT smelyanskiymikhail highperformance3dcompressivesensingmrireconstructionusingmanycorearchitectures
AT haiderclifton highperformance3dcompressivesensingmrireconstructionusingmanycorearchitectures
AT dubeypradeep highperformance3dcompressivesensingmrireconstructionusingmanycorearchitectures
AT manducaarmando highperformance3dcompressivesensingmrireconstructionusingmanycorearchitectures