Quick Start

This page shows example commands to run AutoHS. Install options (Docker, Apptainer, source) are on Installation.

Validate your dataset

AutoHS expects a BIDS-valid dataset with at least one T1-weighted anatomical scan per subject. We recommend validating with the BIDS Validator.

bids-validator /path/to/bids_dataset

Docker example (one subject)

Requires FreeSurfer license.txt when not using --fastsurfer.

docker run --rm \
  -v /path/to/bids_dataset:/data:ro \
  -v /path/to/output:/out \
  -v /path/to/license.txt:/license.txt:ro \
  autohs/autohs:latest \
  /data /out participant \
  --participant-label 001 \
  --fs-license-file /license.txt

Apptainer example (HPC)

Set FREESURFER_SIF / FASTSURFER_SIF before running (see Installation).

export AUTOHS_RUNTIME=apptainer
export FREESURFER_SIF=/path/to/freesurfer_7.4.1.sif

apptainer exec --bind /path/to/bids_dataset:/data:ro \
  --bind /path/to/output:/out \
  --bind /path/to/license.txt:/license.txt:ro \
  autohs.sif \
  /data /out participant \
  --participant-label 001 \
  --fs-license-file /license.txt

FastSurfer (no FreeSurfer license)

python run.py /path/to/bids_dataset /path/to/output participant \
  --participant-label 001 \
  --fastsurfer \
  --runtime apptainer

FastSurfer is typically 5–10× faster than FreeSurfer on CPU. CLI reference: Usage.