/3D-Densenet-Alzheimer

3D Densenet Ensemble applied in 4-way classification of Alzheimer's Disease (BI 2020)

Primary LanguagePythonMIT LicenseMIT

3D Densenet in 4-way Classification of Alzheimer's Disease

Paper published in the Brain Informatics 2020 peer-reviewed journal.
Ruiz J., Mahmud M., Modasshir M., Shamim Kaiser M., Alzheimer’s Disease Neuroimaging Initiative,
"3D DenseNet Ensemble in 4-Way Classification of Alzheimer’s Disease",
Brain Informatics. BI 2020.

Dataset and weights

The dataset and the model weights can be downloaded from this drive.
Both NiFTiFiles and model_weights folders have to be on the same folder as the python code.

Folder structure

.
├── ...
├── cnn_interpretability        
│   └── utils.py
├── model_weights           
│   ├── classifier1_weights.pth         
│   ├── classifier2_weights.pth    
│   └── classifier3_weights.pth
├── NiFTiFiles 
│   └── all the .nii files ...
├── model1.py
├── model2.py
├── model3.py
└── ensemble.py

Dependencies

Developed using CUDA 10.1 with cudnn 8.0.2

Libraries