atlas_arnold's Introduction

Atlas Arnold

This repos allows to perfom the virtual dissection to extract left (L) and right (R) Arnold tracts.

L_Arnold_proper, R_Arnold_proper
L_Arnold_lateral, R_Arnold_lateral
L_Arnold_AR_like, R_Arnold_AR_like
L_Arnold_OR_like, R_Arnold_OR_like

AR := Accoustic Radiations, OR := Optic Radiations

Figure from the paper

How to cite

Maldonado IL, Descoteaux M, Rheault F, et al.
Multimodal study of multilevel pulvino-temporal connections: a new piece in the puzzle of lexical retrieval networks.
Brain. Published online January 18, 2024. doi:10.1093/brain/awae021

Download the paper: Mandonnet et al BRAIN 2024

How to install

The script is based on scilpy

Make sure your pip is up-to-date before trying to install:

pip install --upgrade pip
pip install scilpy==1.5.0

If you have any issues to install scilpy please post an issue here.

Key points to know

- The tractogram needs to be in trk format in **MNI space**.
- The tractogram quality will not be checked. Arnold tracks are "hard-to-track" anatomical pulvino-temporal
  connections. See Mandonnet et al BRAIN 2024 for tractography details. In this work, an aggressive seeding
  TractoFlow pipeline was used in conjunction with a Bundle-Specitifc Tractography (BST) approach. 
- This script does not perform any registration or warping (see scilpy for documentation on this).

How to use

./dissect_arnold_tracks.sh -i INPUT -r MNI_binary_ROIs -o OUT_DIR

  INPUT=/path/to/[INPUT] INPUT folder containing multiple subjects trk in MNI space

                 [INPUT]
                 ├── S1
                 │   └── *.trk
                 ├── S2
                 │   └── *.trk
                 └── *

  MNI_ROI_PATH=/path/to/atlas_arnold/data/MNI_ROIs  Path to MNI_binary_ROIs provided in the atlas_arnold_project

  OUT_DIR=/path/to/[OUT_DIR] Output directory