Making Human Connectome Faster: GPU Acceleration of Brain Network Analysis

Wu, Di; Wu, Tianji; Shan, Yi; Wang, Yu; He, Yong; Xu, Ning; Yang, Huazhong

doi:10.1109/icpads.2010.105

Cited by 7 publications

(6 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our GPU implementation achieves 10 ∼ 43× speedups over the single-core CPU version. The speedups are smaller than those in our previous work [11] for the following two reasons. First, we use double-precision in this work (rather than single-precision in [11]) for higher accuracy.…”

Section: B Performancecontrasting

confidence: 44%

“…The speedups are smaller than those in our previous work [11] for the following two reasons. First, we use double-precision in this work (rather than single-precision in [11]) for higher accuracy. GPU suffers more performance losses from double-precision operations than CPU does.…”

Section: B Performancecontrasting

confidence: 44%

“…GPU suffers more performance losses from double-precision operations than CPU does. Second, the networks in this work is sparser than those used in [11]. From Table III, we can see the speedup is less significant on sparser graphs, where the GPU computing resources are not fully utilized.…”

Section: B Performancementioning

confidence: 86%

“…So we construct a GPU runtime library of these linear algebra algorithms and use CPU to perform the general scheduling. A detailed description can be found in our previous work [11], which is the first GPU implementation of this algorithm.…”

Section: Modular Detection On Gpumentioning

confidence: 99%

“…We use 64 threads (a wavefront in AMD GPU) for one voxelpair correlation. Detailed implementation can be found in our previous work [11].…”

Section: B Correlation Calculation On Gpumentioning

confidence: 99%

See 4 more Smart Citations

A heterogeneous accelerator platform for multi-subject voxel-based brain network analysis

Wang¹,

Xu²,

Ren$^{³

et al. 2011

2011 IEEE/ACM International Conference on Computer-Aided Design (ICCAD)

Self Cite

View full text Add to dashboard Cite

Abstract-The research on understanding the human brain has attracted more and more attention. A promising method is to model the brain as a network based on modern imaging technologies and then to apply graph theory algorithms for analysis. In this work, we examine the computing bottleneck of this method, and propose a CPU-GPU heterogeneous platform to accelerate the process. We construct a statistical brain network from a sample of 198 people and get characteristics such as nodal degree and modularity. This is the first study of voxelbased brain networks on large samples. We also illustrate that domain-specific hardware platform can have a significant impact on neuroscience studies.

show abstract

Section: B Performancecontrasting

confidence: 44%

Section: B Performancecontrasting

confidence: 44%

Section: B Performancementioning

confidence: 86%

Section: Modular Detection On Gpumentioning

confidence: 99%

“…We use 64 threads (a wavefront in AMD GPU) for one voxelpair correlation. Detailed implementation can be found in our previous work [11].…”

Section: B Correlation Calculation On Gpumentioning

confidence: 99%

See 3 more Smart Citations

A heterogeneous accelerator platform for multi-subject voxel-based brain network analysis

Wang¹,

Xu²,

Ren$^{³

et al. 2011

2011 IEEE/ACM International Conference on Computer-Aided Design (ICCAD)

Self Cite

View full text Add to dashboard Cite

show abstract

A Novel Human Body Area Network for Brain Diseases Analysis

Lin

2016

J Med Syst

View full text Add to dashboard Cite

Development of wireless sensor and mobile communication technology provide an unprecedented opportunity for realizing smart and interactive healthcare systems. Designing such systems aims to remotely monitor the health and diagnose the diseases for users. In this paper, we design a novel human body area network for brain diseases analysis, which is named BABDA. Considering the brain is one of the most complex organs in the human body, the BABDA system provides four function modules to ensure the high quality of the analysis result, which includes initial data collection, data correction, data transmission and comprehensive data analysis. The performance evaluation conducted in a realistic environment with several criteria shows the availability and practicability of the BABDA system.

show abstract

Medical image processing on the GPU – Past, present and future

Eklund

Dufort

Forsberg

et al. 2013

Medical Image Analysis

351

198

View full text Add to dashboard Cite

Graphics processing units (GPUs) are used today in a wide range of applications, mainly because they can dramatically accelerate parallel computing, are affordable and energy efficient. In the field of medical imaging, GPUs are in some cases crucial for enabling practical use of computationally demanding algorithms. This review presents the past and present work on GPU accelerated medical image processing, and is meant to serve as an overview and introduction to existing GPU implementations. The review covers GPU acceleration of basic image processing operations (filtering, interpolation, histogram estimation and distance transforms), the most commonly used algorithms in medical imaging (image registration, image segmentation and image denoising) and algorithms that are specific to individual modalities (CT, PET, SPECT, MRI, fMRI, DTI, ultrasound, optical imaging and microscopy). The review ends by highlighting some future possibilities and challenges.

show abstract

Making Human Connectome Faster: GPU Acceleration of Brain Network Analysis

Cited by 7 publications

References 17 publications

A heterogeneous accelerator platform for multi-subject voxel-based brain network analysis

A heterogeneous accelerator platform for multi-subject voxel-based brain network analysis

A Novel Human Body Area Network for Brain Diseases Analysis

Medical image processing on the GPU – Past, present and future

Contact Info

Product

Resources

About