Probabilistic brain fiber tractography on GPUs

Mo Xu; Xiaorui Zhang; Yu Wang; Ling Ren; Ziyu Wen; Yi Xu; Gaolang Gong; Ningyi Xu; Huazhong Yang

doi:10.1109/IPDPSW.2012.92

Probabilistic brain fiber tractography on GPUs

Mo Xu, Xiaorui Zhang, Yu Wang, Ling Ren, Ziyu Wen, Yi Xu, Gaolang Gong, Ningyi Xu, Huazhong Yang

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Diffusion Tensor Magnetic Resonance Imaging (DT-MRI) is an emerging technique that explores the structural connectivity of the human brain. The probabilistic fiber tractography based on DT-MRI data behaves more robustly than deterministic approaches in the presence of fiber crossings, but requires more prohibitive computational time. In this work we present a GPU-based probabilistic framework for brain fiber tractography. The framework includes two main steps: 1) Markov-Chain Monte-Carlo (MCMC) sampling, and 2) probabilistic streamlining fiber tracking. We implement the Metropolis-Hastings sampling for local parameter estimation on GPU. In the probabilistic streamlining fiber tracking, we find that fiber lengths are exponentially distributed, and propose a novel segmenting strategy to improve the load balance. On mid-range GPUs, we achieve performance gains up to 34x and 50x over CPUs for the two steps respectively.

Original language	English (US)
Title of host publication	Proceedings of the 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012
Pages	742-751
Number of pages	10
DOIs	https://doi.org/10.1109/IPDPSW.2012.92
State	Published - Oct 18 2012
Externally published	Yes
Event	2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012 - Shanghai, China Duration: May 21 2012 → May 25 2012

Publication series

Name	Proceedings of the 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012

Other

Other	2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012
Country	China
City	Shanghai
Period	5/21/12 → 5/25/12

Keywords

DT-MRI
GPU
MCMC
Probabilistic Streamlining
Probabilistic fiber tractography

ASJC Scopus subject areas

Software

Access to Document

10.1109/IPDPSW.2012.92

Fingerprint Dive into the research topics of 'Probabilistic brain fiber tractography on GPUs'. Together they form a unique fingerprint.

Cite this

Xu, M., Zhang, X., Wang, Y., Ren, L., Wen, Z., Xu, Y., Gong, G., Xu, N., & Yang, H. (2012). Probabilistic brain fiber tractography on GPUs. In Proceedings of the 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012 (pp. 742-751). [6270714] (Proceedings of the 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012). https://doi.org/10.1109/IPDPSW.2012.92

Probabilistic brain fiber tractography on GPUs. / Xu, Mo; Zhang, Xiaorui; Wang, Yu; Ren, Ling; Wen, Ziyu; Xu, Yi; Gong, Gaolang; Xu, Ningyi; Yang, Huazhong.

Proceedings of the 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012. 2012. p. 742-751 6270714 (Proceedings of the 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Xu, M, Zhang, X, Wang, Y, Ren, L, Wen, Z, Xu, Y, Gong, G, Xu, N & Yang, H 2012, Probabilistic brain fiber tractography on GPUs. in Proceedings of the 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012., 6270714, Proceedings of the 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012, pp. 742-751, 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012, Shanghai, China, 5/21/12. https://doi.org/10.1109/IPDPSW.2012.92

Xu M, Zhang X, Wang Y, Ren L, Wen Z, Xu Y et al. Probabilistic brain fiber tractography on GPUs. In Proceedings of the 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012. 2012. p. 742-751. 6270714. (Proceedings of the 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012). https://doi.org/10.1109/IPDPSW.2012.92

Xu, Mo ; Zhang, Xiaorui ; Wang, Yu ; Ren, Ling ; Wen, Ziyu ; Xu, Yi ; Gong, Gaolang ; Xu, Ningyi ; Yang, Huazhong. / Probabilistic brain fiber tractography on GPUs. Proceedings of the 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012. 2012. pp. 742-751 (Proceedings of the 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012).

@inproceedings{76ac17ef7929475b80640a25289e58d8,

title = "Probabilistic brain fiber tractography on GPUs",

abstract = "Diffusion Tensor Magnetic Resonance Imaging (DT-MRI) is an emerging technique that explores the structural connectivity of the human brain. The probabilistic fiber tractography based on DT-MRI data behaves more robustly than deterministic approaches in the presence of fiber crossings, but requires more prohibitive computational time. In this work we present a GPU-based probabilistic framework for brain fiber tractography. The framework includes two main steps: 1) Markov-Chain Monte-Carlo (MCMC) sampling, and 2) probabilistic streamlining fiber tracking. We implement the Metropolis-Hastings sampling for local parameter estimation on GPU. In the probabilistic streamlining fiber tracking, we find that fiber lengths are exponentially distributed, and propose a novel segmenting strategy to improve the load balance. On mid-range GPUs, we achieve performance gains up to 34x and 50x over CPUs for the two steps respectively.",

keywords = "DT-MRI, GPU, MCMC, Probabilistic Streamlining, Probabilistic fiber tractography",

author = "Mo Xu and Xiaorui Zhang and Yu Wang and Ling Ren and Ziyu Wen and Yi Xu and Gaolang Gong and Ningyi Xu and Huazhong Yang",

year = "2012",

month = oct,

day = "18",

doi = "10.1109/IPDPSW.2012.92",

language = "English (US)",

isbn = "9780769546766",

series = "Proceedings of the 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012",

pages = "742--751",

booktitle = "Proceedings of the 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012",

note = "2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012 ; Conference date: 21-05-2012 Through 25-05-2012",

}

TY - GEN

T1 - Probabilistic brain fiber tractography on GPUs

AU - Xu, Mo

AU - Zhang, Xiaorui

AU - Wang, Yu

AU - Ren, Ling

AU - Wen, Ziyu

AU - Xu, Yi

AU - Gong, Gaolang

AU - Xu, Ningyi

AU - Yang, Huazhong

PY - 2012/10/18

Y1 - 2012/10/18

N2 - Diffusion Tensor Magnetic Resonance Imaging (DT-MRI) is an emerging technique that explores the structural connectivity of the human brain. The probabilistic fiber tractography based on DT-MRI data behaves more robustly than deterministic approaches in the presence of fiber crossings, but requires more prohibitive computational time. In this work we present a GPU-based probabilistic framework for brain fiber tractography. The framework includes two main steps: 1) Markov-Chain Monte-Carlo (MCMC) sampling, and 2) probabilistic streamlining fiber tracking. We implement the Metropolis-Hastings sampling for local parameter estimation on GPU. In the probabilistic streamlining fiber tracking, we find that fiber lengths are exponentially distributed, and propose a novel segmenting strategy to improve the load balance. On mid-range GPUs, we achieve performance gains up to 34x and 50x over CPUs for the two steps respectively.

AB - Diffusion Tensor Magnetic Resonance Imaging (DT-MRI) is an emerging technique that explores the structural connectivity of the human brain. The probabilistic fiber tractography based on DT-MRI data behaves more robustly than deterministic approaches in the presence of fiber crossings, but requires more prohibitive computational time. In this work we present a GPU-based probabilistic framework for brain fiber tractography. The framework includes two main steps: 1) Markov-Chain Monte-Carlo (MCMC) sampling, and 2) probabilistic streamlining fiber tracking. We implement the Metropolis-Hastings sampling for local parameter estimation on GPU. In the probabilistic streamlining fiber tracking, we find that fiber lengths are exponentially distributed, and propose a novel segmenting strategy to improve the load balance. On mid-range GPUs, we achieve performance gains up to 34x and 50x over CPUs for the two steps respectively.

KW - DT-MRI

KW - GPU

KW - MCMC

KW - Probabilistic Streamlining

KW - Probabilistic fiber tractography

UR - http://www.scopus.com/inward/record.url?scp=84867430877&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84867430877&partnerID=8YFLogxK

U2 - 10.1109/IPDPSW.2012.92

DO - 10.1109/IPDPSW.2012.92

M3 - Conference contribution

AN - SCOPUS:84867430877

SN - 9780769546766

T3 - Proceedings of the 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012

SP - 742

EP - 751

BT - Proceedings of the 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012

T2 - 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2012

Y2 - 21 May 2012 through 25 May 2012

ER -

Probabilistic brain fiber tractography on GPUs

Abstract

Publication series

Other

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Cite this