For the classification of the STN-DBS ON/OFF states, the extracted feature maps were organised into a vectorized format reshaping the 3D data into a 1D vector. A mask was later applied to remove the zero values surrounding the brain so that each element in the vectors will represent a specific voxel of the corresponding connectivity map. For each measure, nine classification algorithms were implemented. The default parameters were used for all cases.
-
Clone the repository:
git clone https://github.com/kiakoudimi/NeuroDBS.git
-
Create a conda environment:
conda create -n neuroDBS python=3.10 conda activate neuroDBS
-
Install dependencies:
pip install -r requirements.txt
After setting up your environment and installing dependencies, you can run the main scripts as follows:
Run directly:
cd ../NeuroDBS/scripts/
jupyter notebook
main.ipynbThe dataset is available at IEEE DataPort doi:10.21227/tavd-s033.