# AlzModProgressionAssessment

**Repository Path**: linweichiu/AlzModProgressionAssessment

## Basic Information

- **Project Name**: AlzModProgressionAssessment
- **Description**: A pytorch implementation of multi-modal Alzheimer's disease progression assessment
- **Primary Language**: Python
- **License**: Apache-2.0
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2024-12-25
- **Last Updated**: 2024-12-25

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# AlzModProgressionAssessment

<p align="center">
    <!-- <a href="https://github.com/yyywxk/AlzModProgressionAssessment/blob/main/LICENSE">
        <img alt="license" src="https://img.shields.io/badge/LICENSE-Apache%202.0-blue">
    </a> -->
    <a href="https://github.com/yyywxk/AlzModProgressionAssessment/blob/main/LICENSE">
        <img alt="license" src="https://img.shields.io/github/license/yyywxk/AlzModProgressionAssessment">
    </a>
    <a href="https://github.com/yyywxk/AlzModProgressionAssessment/pulls">
        <img alt="prs" src="https://img.shields.io/github/issues-pr/yyywxk/AlzModProgressionAssessment">
    </a>
    <a href="https://github.com/yyywxk/AlzModProgressionAssessment/issues">
        <img alt="issues" src="https://img.shields.io/github/issues/yyywxk/AlzModProgressionAssessment?color=pink">
    </a>
    <a href="https://github.com/yyywxk/AlzModProgressionAssessment">
        <img alt="issues" src="https://img.shields.io/github/stars/yyywxk/AlzModProgressionAssessment">
    </a>
    <a href="mailto: qiulinwei@buaa.edu.cn">
        <img alt="emal" src="https://img.shields.io/badge/contact_me-email-yellow">
    </a>
</p>

## Overview

This code is the pytorch implementation of the reproduction of two papers:

- Qiu et al's: Development and validation of an interpretable deep learning framework for Alzheimer’s disease classification, published in *Brain* (https://doi.org/10.1093/brain/awaa137).
- Zhang et al's: Alzheimer's disease classification method based on multimodal data, published in [Journal of Computer Applications](http://www.joca.cn/) (https://www.joca.cn/EN/abstract/abstract26300.shtml)

This code follow the setting of [GFE-Mamba](https://github.com/Tinysqua/GFE-Mamba), which can be used to accurately predict whether a patient will process to AD from MCI in a period (symbolled as Δ t) with Multimodal (including MRI, PET and multiple scale assessments) information without the PET engagement. 

## Requirements

- Please clone this repository and navigate to it in your terminal.
- Then prepare an environment with python=3.10, and then use the command `pip install -r requirements.txt` for the dependencies.

## Datasets

- We support two datasets which are from **ANDI** dataset [[Baidu Cloud](https://pan.baidu.com/s/17HFQ3fOrXUvmTYhpnrzWEg?pwd=pbjn)]. More details can be found in [GFE-Mamba: Mamba-based AD Multi-modal Progression Assessment via Generative Feature Extraction from MCI](https://arxiv.org/abs/2407.15719) and its repo [GFE-Mamba](https://github.com/Tinysqua/GFE-Mamba).
- All data are set in `ANDI/` directory.

## Getting Started

To get started with this code, follow these steps:

1. Clone the repository:
   ```bash
   git clone https://github.com/yyywxk/AlzModProgressionAssessment.git
   ```

2. Train and test the Qiu et al's model:

   ```bash
   python train_brain.py
   python test_brain.py
   ```

3. Train and test the Zhang et al's model:

   ```bash
   python train_resnet.py
   python test_resnet.py
   ```

4. To change the configs, please modify:

   ```bash
   ./config/brain.yaml  # for Qiu et al's model
   ./config/resnet.yaml  # for Zhang et al's model
   ```

## Questions

For any questions or feedback, feel free to contact us at [qiulinwei@buaa.edu.cn](mailto:qiulinwei@buaa.edu.cn).

## Acknowledgement

[GFE-Mamba](https://github.com/Tinysqua/GFE-Mamba)

[brain2020](https://github.com/vkola-lab/brain2020)