ModelArray

ModelArray performs mass-univariate statistical analysis of fixel-wise diffusion MRI data to quantify white matter organization and test linear and nonlinear associations.

Key Features:

• Support for statistical models: Implements mass-univariate analyses using linear models and generalized additive models (GAMs) to test linear and nonlinear effects on fixel-wise metrics.
• Fixel-wise diffusion MRI support: Operates on fixel-wise metrics derived from diffusion MRI to enable fiber-specific analyses within voxels for white matter studies.
• Scalability and memory efficiency: Provides a memory-efficient implementation with detailed memory profiling demonstrating minimal memory usage and the ability to handle large datasets on standard personal computers.

Scientific Applications:

• Philadelphia Neurodevelopmental Cohort analysis: Applied to fixel-wise diffusion images from the Philadelphia Neurodevelopmental Cohort (n = 938) to identify nonlinear developmental effects in white matter.
• Lifespan and developmental modeling: Enables detection and characterization of nonlinear white matter changes across the lifespan using GAMs.

Methodology:

Mass-univariate statistical analysis on fixel-wise diffusion MRI metrics using linear models and generalized additive models (GAMs).