Note: This project is still in development and may contain bugs. Please report issues if you encounter problems.
T1Prep: T1 PREProcessing Pipeline (aka PyCAT)
T1Prep is a Python pipeline for preprocessing and segmenting T1-weighted MRI data. It supports:
- Bias-field correction and denoising
- Brain extraction (skull stripping)
- Tissue segmentation (GM, WM, CSF)
- Cortical surface reconstruction and thickness estimation
- Non-linear spatial registration to MNI152 space (CAT12's geodesic shooting)
- Atlas-based ROI extraction
- White matter hyperintensity (WMH/lesion) detection
- BIDS derivatives output naming
Skull-stripping and tissue segmentation run the trained networks of DeepMriPrep and of its brain extraction deepbet, which T1Prep builds on (Lukas Fisch et al., "deepmriprep: Voxel-based Morphometry (VBM) Preprocessing via Deep Neural Networks", https://doi.org/10.48550/arXiv.2408.10656). T1Prep runs them with its own code; neither package has to be installed.
Cortical surface reconstruction uses the cat-surf Python package
(Python bindings to the CAT-Surface C library, shipped as pre-built wheels — no separate
CAT-Surface binaries have to be installed).
For full documentation, CLI usage, Docker instructions, and helper scripts see the GitHub repository.
Installation
pip install T1Prep
pip install "T1Prep[tinygrad]" # optional (Python 3.11+): tinygrad backend, T1Prep --nn-backend tinygrad
Multiple Python versions? Install with the exact interpreter you will run T1Prep with, e.g.
python3.12 -m pip install T1Prep(requires Python 3.9–3.14; on macOS prefer 3.10+, since a 3.9 install is pinned to PyTorch 2.8 and runs slower). If the wrong interpreter is auto-detected, point T1Prep at the right one withT1Prep --python /path/to/python …orexport T1PREP_PYTHON=/path/to/python.pipx install T1Prepor a dedicated venv avoids the ambiguity.
pip install places every entry point into the active environment's bin/
directory. With that directory on your PATH (e.g. an activated venv) the
following commands are available:
| Command | Role |
|---|---|
T1Prep |
main CLI — batch + parallel processing (--multi) |
PyCAT |
T1Prep under a second name; identical, only the banner differs |
t1prep-gui, T1Prep_gui |
desktop launcher: choose images, run, browse the results |
t1prep-run |
single-subject Python entry |
CAT_SurfView |
surface viewer |
CAT_VolView |
volume viewer (3 orthogonal slices) |
t1prep-make-apps |
macOS: build .app bundles for the viewers |
t1prep-download-models |
fetch model weights |
t1prep-bbreg |
BOLD reference, head-motion correction and boundary-based BOLD-to-T1w registration for fMRIPrep to reuse |
CAT_PlotHistogram |
histograms of volumes, surfaces or text data |
CAT_PlotBoxplot |
box and violin plots of volumes, surfaces or text data |
CAT_VolDiff, CAT_VolSmooth_ui, CAT_SurfResampleMulti_ui, CAT_SurfParameters_ui, CAT_Surf2ROIMulti_ui, CAT_GrepJson |
command-line helpers for post-processing and reports |
process_longitudinal.sh, realign_longitudinal.sh, warp_longitudinal.sh, modulate_longitudinal.sh |
experimental longitudinal pipeline |
Download model weights
Model weights are not bundled in the wheel (about 250 MB once downloaded). Download them after installation:
t1prep-download-models
Models are stored in your user cache, in a folder named after the model release
(~/Library/Caches/t1prep/models/<release> on macOS, ~/.cache/t1prep/models/<release>
on Linux; T1PREP_MODEL_DIR overrides it) and are downloaded
automatically on first pipeline use if this step is skipped.
Requirements
- Python 3.9–3.14 (3.10+ recommended). PyTorch 2.9+ (Python 3.10+) adds MPS kernels that give a measurable speed-up on Apple Silicon (2:39 min vs 3:02 min per subject); a Python 3.9 install is pinned to PyTorch 2.8 and falls back to the CPU for those kernels. On Linux/CUDA/CPU both stacks perform the same.
- ~250 MB disk space for model weights (downloaded separately, see above)
- For GPU acceleration: CUDA-capable GPU or Apple Silicon (MPS)
Python API
from t1prep import run_t1prep
# Single file — results saved next to input
run_t1prep("/data/sub-01_T1w.nii.gz")
# Single file, BIDS-compatible output
run_t1prep("/data/sub-01/ses-1/anat/sub-01_ses-1_T1w.nii.gz", bids=True)
# Batch processing with options
run_t1prep(
["/data/T1/sub-01.nii.gz", "/data/T1/sub-02.nii.gz"],
out_dir="/results",
atlas=["neuromorphometrics", "suit"],
wp=True, # save warped segmentations
p=True, # save native segmentations
csf=True, # save CSF segmentation
lesions=True, # save WMH lesion map
gz=True, # compress outputs (.nii.gz)
)
run_t1prep processes the files one after another; for parallel batches use
T1Prep --multi N or call it from several processes.
Key parameters
| Parameter | Type | Description |
|---|---|---|
inputs |
str or list[str] |
Input NIfTI file(s) (first argument) |
out_dir |
str |
Output directory (default: same as input) |
atlas |
list[str] |
Atlas names for ROI extraction |
no_surf |
bool |
Skip cortical surface estimation (default: False) |
bids |
bool |
Use BIDS derivatives naming (implies gz) |
gz |
bool |
Save compressed NIfTI (.nii.gz) |
wp |
bool |
Save warped (MNI space) segmentations |
p |
bool |
Save native space segmentations |
csf |
bool |
Save CSF segmentation |
lesions |
bool |
Save WMH/lesion map |
amap |
bool |
Refine the network segmentation with CAT12's AMAP |
skullstrip_only |
bool |
Only run skull stripping then exit |
skip_skullstrip |
bool |
Skip skull stripping (pre-stripped input) |
track |
bool |
Report a crash of the segmentation (default: True) |
Output structure
Non-BIDS (default): subfolders mri/, surf/, report/, label/ in the output directory,
with CAT12-compatible filenames (e.g., mwp1sub-01.nii, lh.thickness.sub-01).
BIDS input (the image lies in an anat/ folder): the results go to
<out_dir or dataset root>/derivatives/T1Prep-v<version>/sub-XX[/ses-YY]/anat/.
bids=True names the files by the BIDS derivatives convention; for an image outside an
anat/ folder they then go directly into the output directory, without the subfolders.
Telemetry
Unless you opt out, T1Prep tells its developers how many files a run
processed and how many failed, and where in the code a segmentation crashed.
File names, paths, images and host names are never sent. --no-track,
run_t1prep(..., track=False), or T1PREP_NO_TRACK=1 / DO_NOT_TRACK=1 in
the environment switch it off; details in
docs/usage.md.
Documentation
| Document | Covers |
|---|---|
| Installation | pip, source checkout, WSL, manual install, Docker |
| Usage | Options, output structure, naming, examples |
| Viewers | CAT_SurfView and CAT_VolView, interactive and batch |
| Tools | Launcher, plot tools and the post-processing helpers |
License
Distributed under the Apache License 2.0.
Metadata
Release files for T1Prep 0.9.1
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| t1prep-0.9.1.tar.gz | 76.4 MB | With provenance | Details for t1prep-0.9.1.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| t1prep-0.9.1-py3-none-any.whl | Python 3 | none | any | Details for t1prep-0.9.1-py3-none-any.whl |
Total release size: 152.4 MB
Release files / t1prep-0.9.1.tar.gz
| Download URL | t1prep-0.9.1.tar.gz |
|---|---|
| Size | 76.4 MB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
a60957df33ba2898d300c9de5e28088168846a2e66ac5d2c869b6c23170a8ec0
|
|
BLAKE2b-256 checksum How to use checksums |
821e34385632ecfb01e5c0ff85f989b4236480e7514c2b8ad5f2ab046bbefa51
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.16
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Nov 19, 2026.
Transparency logRelease files / t1prep-0.9.1-py3-none-any.whl
| Download URL | t1prep-0.9.1-py3-none-any.whl |
|---|---|
| Size | 76.0 MB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
99612e254e0bd830c3738a603236fcda00687746a9942b0bbf3effcfc31f4c95
|
|
BLAKE2b-256 checksum How to use checksums |
d3848b2e70171931cd1f3309a9a5c849f45dd7b1656807219ffeba853078597c
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.16
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Nov 19, 2026.
Transparency logSecurity for T1Prep 0.9.1
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
Full provenance
Both files in this release have been uploaded with provenance metadata.
Attestation for t1prep-0.9.1.tar.gz
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Nov 19, 2026.
Transparency logA note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.