DeepT1-WMH

White Matter Hypointensities lesions for T1-weighted MRI images.

It relies on a Convolutional Neural Network pre-trained on FLAIR segmentations using the large JPSC-AD cohort.

For background and technical details about its creation, refers to this corresponding Human Brain Mapping manuscript: http://doi.org/10.1002/hbm.25899

Requirement

This program requires Python 3, with the PyTorch library, version > 1.4.0., although >=1.7 is highly recommended for improved memory usage

No GPU is required

Installation

Just clone or download this repository.

In addition to PyTorch, the code requires scipy and nibabel.

If not pre-installed, a simple way to install python3 from scratch may be to use a Anaconda (anaconda.com) environment then

• install scipy (conda install scipy or pip install scipy) and nibabel (pip install nibabel)
• get pytorch for Python/CPU from https://pytorch.org/get-started/locally/. CUDA is not necessary.

Usage:

To use the program, simply call:

./deepwmh.sh t1_image.nii.gz

(or it can be added to your PATH)

To process multiple subjects, pass them as multiple arguments. deepwmh.sh subject_*.nii.gz.

The resulting WMH segmentation mask will be named t1_image_mask_wmh.nii.gz, and t1_image_mask_ROIs.nii.gz for the region labels (periventricular, deep-white, infracortical). The lesion total and regional volumes statistics are available in t1_image_wmh_in_lrois.csv.

If multiple input images were specified, a summary table is generated as all_subjects_wmh_report.csv

Optionally, adding "-v" (verbose) in the command line will output more images, including the non-thresholded (probabilistic, 0-255) WMH-lesion segmentation output, named t1_image_prob_wmh.nii.gz .

Issues

If the following message appears:

wgrid = self.grid @ self.tA
                  ^
SyntaxError: invalid syntax

It probably means that the system python was found. Make sure python 3 is in your path or edit deepwmh.sh with a suitable path.