BRAIN-To MRI Protocols Repository

This is a repository of wiki, code and containers for processing data acquired using the BRAIN-TO imaging protocols on Siemens MAGNETOM Prisma 3 T (XA30).

Please refer the article titled, Advancing Clinical and Neuroscientific Research Through Accessible and Optimized Protocol Design at 3T, published in the RSNA Edition of Siemens MAGNETOM Flash (2023) for an overview.

Download Links for BRAIN-TO Protocols

1. Complete Set
   https://zenodo.org/records/10685481

2. Anatomy Set
   https://zenodo.org/records/10685475

3. fMRI Set
   https://zenodo.org/records/10685458

4. Diffusion Set
   https://zenodo.org/records/10685449

5. ASL Set
   https://zenodo.org/records/10685308

This repository is being updated steadily and include the following

1. Zenodo links for XA30 exar1 files
2. Zenodo links for Protocol PDFs (for translating to other Prisma software versions)
3. Wiki with recommended pipelines and tools
4. Adding scripts and example code for processing data

Please feel free to contribute to our efforts.

For urgent requests, please e-mail.

Contributors

Sriranga Kashyap, BRAIN-To Lab, Krembil Brain Institute, UHN
Yuexin Xi, PhD Student, Dept. of Medical Biophysics, University of Toronto

---

bto-docker

General-purpose processing container for BTO protocols.

---

Included tools

Tool	Version
FSL	6.0.7.22
FreeSurfer	8.2.0
oxasl	custom
ANTs	2.6.5
AFNI	latest
MRtrix3	latest

---

Quick start

docker pull brainto/bto-docker:latest

Mount your data directory to /data:

docker run --rm \
    --user $(id -u):$(id -g) \
    --cpuset-cpus "0-9" \
    -v /path/to/data:/data \
    brainto/bto-docker \
    bto-asl --help

FreeSurfer requires a valid license file mounted at runtime:

docker run --rm \
    -v /path/to/data:/data \
    -v /path/to/license.txt:/opt/fs-license/license.txt \
    brainto/bto-docker \
    bto-anat freesurfer --help

All file paths must be absolute paths as seen from inside the container (i.e. under /data).

---

CLI commands

Command	Purpose
bto-asl	ASL perfusion pipeline
bto-anat fsl	FSL structural preprocessing

bto-asl — ASL perfusion pipeline

End-to-end pipeline.

docker run --rm -it \
    --user $(id -u):$(id -g) \
    --cpuset-cpus "0-9" \
    -v $PWD:/data \
    brainto/bto-docker bto-asl \
    --subjid  sub-01_ses-rest \
    --asl     /data/sub-01/ses-rest/perf/sub-01_ses-rest_asl.nii.gz \
    --t1      /data/sub-01/ses-anat/anat/sub-01_T1w.nii.gz \
    --m0-fwd  /data/sub-01/ses-rest/fmap/sub-01_ses-rest_dir-PA_m0scan.nii.gz \
    --m0-rev  /data/sub-01/ses-rest/fmap/sub-01_ses-rest_dir-AP_m0scan.nii.gz

JSON sidecars (.json alongside each NIfTI) are auto-detected for acquisition parameters. Pass --no-json to skip sidecar detection and use assumed PE direction.

If structural preprocessing has already been run, pass --anatdir to reuse the existing .anat directory:

    --anatdir /data/sub-01/ses-anat/anat/sub-01_T1w.anat

Key flags

Flag	Default	Description
--subjid	required	Subject ID — names the output directory bto-asl_<subjid>/
--asl	required	4D ASL NIfTI (repeat for multiple runs)
--t1	—	T1w structural NIfTI — triggers FSL structural preprocessing
--anatdir	—	Existing .anat directory — skips structural preprocessing
--m0-fwd	—	Forward-PE M0 calibration image
--m0-rev	—	Reverse-PE M0 for topup distortion correction
--outdir	alongside --asl	Base output directory
--no-json	off	Skip JSON sidecar requirement
--pe-dir	j	Forward PE direction when --no-json is active

---

bto-anat — structural preprocessing

FSL

docker run --rm -v /path/to/data:/data \
    brainto/bto-docker bto-anat fsl \
    --t1 /data/sub-01/anat/T1w.nii.gz

Runs bias correction, brain extraction (mri_synthstrip), linear and nonlinear MNI registration, FAST tissue segmentation, and subcortical segmentation. Output defaults to <t1_parent>/derivatives/anat/<t1_stem>.anat/.

Notes

• FreeSurfer license is never baked into the image — mount it at runtime to /opt/fs-license/license.txt
• Base image: ubuntu:24.04
• All scripts use FSL's conda Python (/usr/local/fsl/bin/python3) as the runtime interpreter
• Default Docker runs as root; pass --user $(id -u):$(id -g) to write output as the current user

---

Dev

github.com/srikash/bto-docker