Developing a CNN-based classfier for identifying connectomes from subjects with ASD.
CHARM
dataset provided by collaborators at Penn Medicine- Connectomes from 450 subjects in total, 313 of which passed Data QA check
- 163 ASD (Autism-Spectrum Disorder) & 150 TDC (Typically Developing Controls)
- 2 Sets of connectomes:
- Desikan atlas with 86 parcellations
- Schaefer atlas with 220 parcellations
# Clone the repository
git clone ...
cd ...
# Set up conda environment
conda install environment.yml
# Clean up demographics table
python