ProbioticFarmer/mindboggle-101-ants

Mindboggle-101 — Complete Lab-Processed ANTs antsCorticalThickness & SyN Warp Release

Complete lab-computed antsCorticalThickness.sh outputs over the Mindboggle-101 cohorts (Klein & Tourville, 2012; Zenodo 10.5281/zenodo.22071936, CC-BY 4.0), covering 140 subject-sessions — including 20 paired baseline ↔ rescan OASIS-TRT-20 diffeomorphic registration & Atropos tissue-posterior bundles.

Key Volumetric & Diffeomorphic Assets per Subject (cohorts/{cohort}/{subject_id}/)

• *SubjectToTemplate1Warp.nii.gz & *TemplateToSubject0Warp.nii.gz — Full 3D nonlinear ANTs SyN diffeomorphic displacement fields $\hat{u}(x)$.
• *SubjectToTemplateLogJacobian.nii.gz — Log-Jacobian determinant maps of the SyN warp.
• *BrainSegmentationPosteriors1..6.nii.gz — Atropos 6-class posterior probability maps (1=CSF, 2=Cortical GM $P_{\text{GM}}(x)$, 3=WM, 4=Deep GM, 5=Brainstem, 6=Cerebellum).
• *BrainExtractionBrain.nii.gz & *BrainSegmentationN4.nii.gz — Brain-extracted & N4 bias-corrected 1mm T1w volumes.
• *CorticalThickness.nii.gz & *CorticalThicknessNormalizedToTemplate.nii.gz — DiReCT cortical thickness maps in native and template space.

Fast Usage on Hugging Face Jobs

from huggingface_hub import snapshot_download
import pandas as pd
import nibabel as nib

# Pull metadata + OASIS-TRT-20 baseline & rescan ANTs SyN warps + GM posteriors
root = snapshot_download(
    repo_id="ProbioticFarmer/mindboggle-101-ants",
    repo_type="dataset",
    allow_patterns=["metadata/*", "cohorts/OASIS-TRT-20*/*"],
)
pairs = pd.read_parquet(f"{root}/metadata/oasis_trt_pairs.parquet")
warp_img = nib.load(f"{root}/{pairs.loc[0, 'baseline_syn_warp_path']}")
gm_post_img = nib.load(f"{root}/{pairs.loc[0, 'baseline_gm_posterior_path']}")

Citation

Klein, A., & Tourville, J. (2012). 101 labeled brain images and a consistent human cortical labeling protocol. Frontiers in Neuroscience, 6, 171. (doi:10.5281/zenodo.22071936)