In this directory, I made available the code used to train the age prediction model of the article "Longitudinal analysis of AI predicted brain age in amnestic and non-amnestic sporadic early-onset Alzheimer's disease".
The code is divided into five files :
- callbacks.py - which contains the custom callbacks functions ;
- data_agumentation.py - which contains data augmentation functions on three dimensional images;
- generators.py - which contains the generators able to create our batch of MRI images;
- regression.py - which contains the architecture of the model used;
- brain_age.ipynb - the notebook that uses each of the above functions to train the age prediction model.