ATPP: A Pipeline for Automatic Tractography-Based Brain Parcellation.

Area of Science:

• Neuroimaging
• Computational Neuroscience
• Brain Atlasing

Background:

• Brain parcellation efforts aim to delineate distinct brain areas using micro-structural, macro-structural, or connectional features.
• Connectivity-based parcellation, leveraging multimodal magnetic resonance imaging (MRI), has gained prominence.
• Existing brain atlases, like the Brainnetome Atlas, utilize this framework but face challenges with large datasets and computational demands.

Purpose of the Study:

• To present an integrated, open-source pipeline for automated brain parcellation.
• To address the computational challenges and lack of dedicated tools in connectivity-based brain atlasing.
• To facilitate the creation and utilization of detailed brain atlases.

Main Methods:

• Development of the Automatic Tractography-based Parcellation Pipeline (ATPP).
• Implementation of both command-line and graphical user interface versions for flexibility.
• Utilizing automatic processing and massive parallel computing for efficient parcellation.
• Demonstration on left precentral gyrus and middle frontal gyrus across independent datasets.

Main Results:

• ATPP successfully automates the connectivity-based parcellation framework.
• Both command-line and GUI versions are functional and validated.
• The pipeline efficiently processes high-resolution multimodal MRI data.
• Successful parcellation across diverse brain regions and the Human Brainnetome Atlas.

Conclusions:

• ATPP significantly facilitates brain parcellation, overcoming previous computational and accessibility hurdles.
• The open-source pipeline offers a powerful and flexible solution for researchers.
• ATPP's validation across multiple datasets and brain regions underscores its utility and robustness.