Related Experiment Video
Updated: Jul 13, 2026
08:33
A Randomized, Sham-Controlled Trial of Cranial Electrical Stimulation for Fibromyalgia Pain and Physical Function, Using Brain Imaging Biomarkers
Published on: January 5, 2024
Power and sample size calculation for neuroimaging studies by non-central random field theory
Satoru Hayasaka1, Ann M Peiffer, Christina E Hugenschmidt
1Biostatistical Sciences, Wake Forest University School of Medicine, Winston-Salem, NC 27157, USA. shayasak@wfubmc.edu
Neuroimage
|July 31, 2007
Summary
This study introduces a novel Random Field Theory (RFT) method for power and sample size calculations in neuroimaging. The approach accurately estimates statistical power, aiding in planning brain imaging studies like fMRI.
Area of Science:
- Neuroimaging
- Statistical Analysis
- Brain Imaging
Background:
- Neuroimaging studies face challenges in power and sample size determination due to massive multiple comparisons across numerous correlated voxels.
- Accurate statistical power is crucial for reliable detection of brain signals and effective study design.
Purpose of the Study:
- To develop and validate a novel power analysis method for neuroimaging studies.
- To address the complexities of multiple comparisons and voxel correlations in brain imaging.
- To facilitate accurate estimation of statistical power and sample size requirements.
Main Methods:
- Proposed a power analysis method based on Random Field Theory (RFT).
- Modeled signal areas in neuroimaging data as non-central random fields.
- Validated the framework using Monte-Carlo simulations and applied it to blood oxygenation level dependent (BOLD) functional magnetic resonance imaging (fMRI) data.
Main Results:
- The proposed non-central RFT framework accurately estimates statistical power for specific brain regions.
- The method accounts for the 3D nature of brain signals and can generate power and sample size maps.
- Simulations confirmed the accuracy of power estimation, and application to fMRI data demonstrated its utility in study planning, even with small sample sizes.
Conclusions:
- The novel non-central RFT method provides an accurate and effective approach for power and sample size calculations in neuroimaging.
- This framework enhances the planning and sensitivity analysis of brain imaging studies, including fMRI.
- The method aids researchers in determining the necessary number of subjects to reliably detect signals.
