Skip to content

Latest commit

 

History

History
8 lines (6 loc) · 767 Bytes

README.md

File metadata and controls

8 lines (6 loc) · 767 Bytes

Neurodegeneration prediction using MRI imaging data

This is the codebase for paper titled "Prediction, prognosis and monitoring of neurodegeneration at biobank-scale via machine learning and imaging".

Key points:

  • MRI features were extracted using a seperate open source tool (ANTsPyT1w: https://github.com/stnava/ANTsPyT1w).
  • In the above code, the link to preprocessed data files are left empty due to protect data privacy. Information of related files are described in comments for better code readability.
  • All the compute / model training were performed using NIH biowulf (https://hpc.nih.gov/). So, we did not include the codes designed specifically to submit jobs on biowulf. Details of model training and optimal parameters are described in the paper.