A Unified Framework for Multi-Site and Multi-Modal Alzheimer’s Disease Diagnosis on MRI

This repo contains PyTorch implementation of the paper: A Unified Framework for Multi-Site and Multi-Modal Alzheimer’s Disease Diagnosis on MRI

Paper Link

Setup Instructions

Datasets

Directory structure

├── Data
 ├── AD01
 │   ├── 1_NC001
 │   │   ├── T1_dwi.nii.gz
 │   │   ├── md.nii.gz
 │   │   ├──data_md_128.pth
 │   ├── 1_NC002
 │   │   ├── T1_dwi.nii.gz
 │   │   ├── md.nii.gz
 │   │   ├──data_md_128.pth
 │   ├── ...
 ├── AD02
 ├── AD03
 ├── ...

Dependancies

pip install -r ./requirements.txt

Run the Model

Preprocessing：

python dataset/pre_process.py --target_shape "(128, 128, 128)" --gamma_range "(0.5, 1.5)" --root_dir "data/" --destination_dir "data/" --file_name "data_128.pth"

Train：

python train.py

Eval:

python eval.py

References

If you think our research work helpful, please consider citing our original paper.