compneuro-atrophymapping

This repo contains the code used by the Computational Neuroimaging Lab at BioBizkaia HRI to compute the atrophy w-maps using T1w images, aimed to be used in network mapping analyses.

Pipeline Description

Prerequisites

NOTE: SECTION SUBJECT TO CHANGE In case you have not run VBM yet:

• T1w images of a control group
• T1w images of a clinical group This tool will run VBM on the provided images first.

In case you have already run the VBM pipelines on each study group, you could use the following files:

• GM_mod_merg_sX.nii.gz of a control group
• GM_mod_merg_sX.nii.gz of a clinical group

If not all groups have GM_mod_merg_sX, this tool will detect which groups have it and which do not, and will run the VBM pipeline for the groups that do not have it.

Getting Started

Usage

Computing W-score maps

Computes the W-score maps of the study group with respect to the control group. compute_wmaps [-h] [-vbmcn CONTROL_VBM_PATH] [-vbmpat STUDYGROUP_VBM_PATH] [-t1cn CONTROL_T1_DIR] [-t1pat STUDYGROUP_T1_DIR] [-dcn CONTROL_DESIGN_MATRIX] [-dpat STUDYGROUP_DESIGN_MATRIX] [-n N_JOBS] -o OUTPUT_DIR

Binarizing W-score maps

Binarizes the W-score maps using a threshold. The threshold is applied to negative and positive values separately, producing atrophy (< -threshold) and growth (> +threshold) maps, respectively. binarize_wmaps [-h] --wmap WMAP --threshold THRESHOLD

Filtering clusters in binarized W-score maps

Removes binary clusters that are smaller than a given percentile, respect to the biggest cluster in the map. filter_clusters [-h] --clusterim CLUSTERIM --percentile PERCENTILE [--merge MERGE]

Outputs

TO BE WRITTEN

Citing

The methodology followed in this tool is based on the following work: Ossenkoppele, R., Cohn-Sheehy, B.I., La Joie, R., Vogel, J.W., Möller, C., Lehmann, M., van Berckel, B.N.M., Seeley, W.W., Pijnenburg, Y.A., Gorno-Tempini, M.L., Kramer, J.H., Barkhof, F., Rosen, H.J., van der Flier, W.M., Jagust, W.J., Miller, B.L., Scheltens, P. and Rabinovici, G.D. (2015), Atrophy patterns in early clinical stages across distinct phenotypes of Alzheimer's disease. Hum. Brain Mapp., 36: 4421-4437.

We thank Rik Ossenkoppele and Colin Groot-Remmers for sharing their original code and their insights, which helped us to develop this tool.

License

This project is licensed under the Apache License 2.0 - see the LICENSE file for details

TODO (highest priority first)

• Adapt the code to work with joint VBM analysis, not per group (new branch)
• Add a section on how to install
• Dockerize