BrainSuite BIDS App: Containerized Workflows for MRI Analysis

Area of Science:

• Neuroimaging
• Computational Neuroscience
• Medical Image Analysis

Background:

• Growing need for standardized computational methods in neuroimaging to ensure reproducibility and portability.
• The Brain Imaging Data Structure (BIDS) provides a standard for organizing neuroimaging data.
• Containerized processing environments (BIDS Apps) streamline the execution of complex analysis workflows.

Purpose of the Study:

• To present the BrainSuite BIDS App, integrating BrainSuite's MRI processing capabilities into the BIDS App framework.
• To provide standardized participant-level and group-level analysis workflows for structural, diffusion, and functional MRI data.
• To introduce the BrainSuite Dashboard for real-time quality control of neuroimaging analysis pipelines.

Main Methods:

• Development of a BrainSuite BIDS App encompassing three participant-level pipelines: Anatomical, Diffusion, and Functional.
• Implementation of group-level analysis workflows utilizing BrainSuite Statistics Toolbox (bstr) and BrainSync.
• Integration of the BrainSuite Dashboard for interactive quality control and parameter adjustment.

Main Results:

• The BrainSuite BIDS App successfully encapsulates core BrainSuite MRI processing functionalities within the BIDS App standard.
• Demonstrated participant-level processing of T1-weighted MRI, diffusion-weighted imaging (DWI), and functional MRI (fMRI) data.
• Validated group-level analysis capabilities and real-time quality control through the BrainSuite Dashboard.

Conclusions:

• The BrainSuite BIDS App offers a robust and standardized mechanism for deploying BrainSuite workflows in diverse research settings.
• Facilitates large-scale neuroimaging studies by enhancing data processing reproducibility, portability, and quality assurance.
• Enables efficient analysis of structural, diffusion, and functional MRI data, contributing to advancements in brain imaging research.