MAD-Former: A Traceable Interpretability Model for Alzheimer’s Disease Recognition based on Multi-patch Attention

python train.py
The interpretability module IAC can be reproduced according to Section III of the manuscript.

The supplementary material provide the significant brain tissues in the 3D ROI with a patch size of 8, as well as the subject IDs used in the manuscript experiments.
2024.1.22 Supplemental generalization evaluation and schematic diagram of the 3D BFEN structure.

@article{ye2024mad,
title={MAD-Former: A Traceable Interpretability Model for Alzheimer's Disease Recognition Based on Multi-Patch Attention},
author={Ye, Jiayu and Zeng, An and Pan, Dan and Zhang, Yiqun and Zhao, Jingliang and Chen, Qiuping and Liu, Yang},
journal={IEEE Journal of Biomedical and Health Informatics},
year={2024},
publisher={IEEE}
}