Mindboggle

Mindboggle analyzes preprocessed T1-weighted MRI data to compute volumetric, surface-based, and tabular morphometric measures for brain morphometry studies.

Key Features:

• Output types: Produces volumetric, surface-based, and tabular data enriched with label, feature, and shape information.
• Comprehensive shape measures: Computes area, volume, cortical thickness, curvature, depth, Laplace-Beltrami spectra, and Zernike moments as morphometric descriptors.
• Large-scale data handling: Capable of processing large datasets and has been applied in extensive population-scale shape analyses.
• Comparative evaluation: Algorithms evaluated against a comprehensive set of manually labeled brain images, representing the largest publicly available dataset of that type and benchmarked against state-of-the-art methods where applicable.

Scientific Applications:

• Biomarker identification for mental health disorders: Enables extraction of morphometric markers used for diagnosis, tracking, and prediction of disorder progression.
• Detection of subtle shape changes: Provides shape measures beyond volume and cortical thickness to explore sensitive and specific structural biomarkers.
• Population-scale morphometry: Supports large-cohort studies investigating structural variability across populations.

Methodology:

Processes preprocessed T1-weighted MRI data to extract volumetric and surface-based morphometric features (including area, volume, thickness, curvature, depth, Laplace-Beltrami spectra, and Zernike moments) and evaluates algorithms against a large set of manually labeled brain images; results, code, and datasets have been made publicly available.