Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Fast-GPU-PCC: A GPU-Based Technique to Compute Pairwise Pearson's Correlation Coefficients for Time Series Data-fMRI

Taban Eslami1, Fahad Saeed2

  • 1Department of Computer Science, Western Michigan University, Kalamazoo, MI 49008, USA. taban.eslami@wmich.edu.

High-Throughput
|April 21, 2018
PubMed
Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Predicting progression of Alzheimer's disease using blood-based multi-omics data.

Bioinformatics advances·2026
Same author

FiCOPS: Hardware/Software Co-Design of FPGA Computational Framework for Mass Spectrometry-Based Peptide Database Search.

bioRxiv : the preprint server for biology·2026
Same author

TITAN-BBB: Predicting BBB Permeability using Multi-Modal Deep-Learning Models.

bioRxiv : the preprint server for biology·2026
Same author

A Machine Learning and Benchmarking Approach for Molecular Formula Assignment of Ultra High-Resolution Mass Spectrometry Data from Complex Mixtures.

bioRxiv : the preprint server for biology·2026
Same author

MolDeBERTa: Foundational Model for Physicochemical and Structural-Informed Molecular Representation Learning.

bioRxiv : the preprint server for biology·2026
Same author

Habenula alterations in resting state functional connectivity among autistic individuals.

bioRxiv : the preprint server for biology·2025

This study introduces Fast-GPU-PCC, a novel graphics processing unit (GPU) algorithm that significantly accelerates the computation of Pearson's correlation coefficients for functional magnetic resonance imaging (fMRI) data, enabling faster brain network analysis.

Area of Science:

  • Neuroimaging
  • Computational Neuroscience
  • Biomedical Engineering

Background:

  • Functional magnetic resonance imaging (fMRI) is crucial for studying brain activity and connectivity.
  • Pearson's correlation coefficient is a standard metric for analyzing functional brain networks.
  • Traditional CPU-based correlation computation is time-consuming for large fMRI datasets.

Purpose of the Study:

  • To develop a GPU-based algorithm for rapid computation of pairwise Pearson's correlation coefficients.
  • To optimize the calculation of functional connectivity in fMRI data.
  • To address the computational bottleneck in analyzing large-scale neuroimaging studies.

Main Methods:

  • Proposed Fast-GPU-PCC algorithm utilizing graphics processing unit (GPU) acceleration.
Keywords:
CUDAGPUPearson’s correlation coefficientfMRImatrix multiplication

Related Experiment Videos

  • Leveraged the symmetric property of Pearson's correlation to compute unique coefficients.
  • Stored computed correlations in an efficient one-dimensional array format.

    • Main Results:

    • Fast-GPU-PCC demonstrated superior performance compared to CPU-based and existing GPU-based methods.
    • The algorithm achieved a speedup of 62x over the CPU version.
    • It was 2-3x faster than other state-of-the-art GPU techniques on real and synthetic fMRI data.

    Conclusions:

    • Fast-GPU-PCC offers a significant computational advantage for fMRI data analysis.
    • The method enables more efficient construction and analysis of brain functional networks.
    • This advancement can accelerate research into brain disorders and functional connectivity.