Early progression detection from MCI to AD using multi-view MRI for enhance assisted living

##This is a Keras implementation of the study conducted for the early detection of progression from MCI to PMCI/AD.

Please refer to the requirements.txt file to install all necessary libraries for running this code.

Training:

Run the following command to train any desired CNN backbone for feature extraction with a Bayesian optimized classification head. You can also use a CBAM module as an intermediate attention module by specifying a flag in the given command.

python train.py --cnnbb [EfficientNet, ResNet, ConvNext, DenseNet, XceptionNet] ResNet --ch [mlp, lstm, multihead_attention] multihead_attention --cbam [True, False] --data_dir "path\to\root\folder"