8. Analyzing fMRI using GLMs

Nilearn’s stats module uses Generalized Linear Model to analyze functional MRI data. Nilearn offers functions to set up first- and second-level models, perform t-tests, multiple-comparisons correction and thresholding, and allows for the generation of html reports that detail the analyses performed and present thresholded results.

• 8.1. An introduction to GLMs in fMRI statistical analysis
  • 8.1.1. A primer on BOLD-fMRI data analysis
    • 8.1.1.1. What is fMRI ?
    • 8.1.1.2. fMRI data modeling
    • 8.1.1.3. fMRI statistical analysis
    • 8.1.1.4. Multiple Comparisons
• 8.2. First level models
  • 8.2.1. HRF models
  • 8.2.2. Design matrix: event-based and time series-based
    • 8.2.2.1. Event-based
    • 8.2.2.2. Time series-based
  • 8.2.3. Fitting a first level model
    • 8.2.3.1. Computing contrasts
  • 8.2.4. Extracting predicted time series and residuals
  • 8.2.5. Surface-based analysis
• 8.3. Second level models
  • 8.3.1. Fitting a second level model
  • 8.3.2. Thresholding statistical maps
  • 8.3.3. Multiple comparisons correction
  • 8.3.4. Voxel based morphometry
• 8.4. Difference in meanings between different toolboxes
  • 8.4.1. Statistical Parametric Mapping (SPM)
  • 8.4.2. FMRIB Software Library (FSL)