Skip to content

Justin-12138/AD-Sliced-Classification

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

26 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

基于图神经网络的AD分类研究

Update

  • 2024.03.20:尝试重新使用FREESURFER处理脑部数据

  • 2024.02.02:完整的workflow.sh🏮

  • 2024.01.28:resNet18(acc:0.7)🚡

  • 2024.01.22:数据处理流程化,尝试已实现方法🍈

  • 2024.01.15:背景调查,以及数据预处理📓

TODO

  • 基本的数据预处理

  • 常见的神经网络模型

    • Multilayer_Perceptron

    • resNet

    • LeNet

    • AlexNet

    • VGG

  • 图神经网络

Tools

python==3.10.8
itksnap-4.0.2-20230925-Linux-gcc64
fsl python2

Dataset

git clone git@github.com:Justin-12138/Gradesign.git
cd Gradesign/ADNI
chmod +x workflow.sh
./workflow.sh
  • 数据集介绍

    https://adni.loni.usc.edu/methods/mri-tool/mri-analysis/#mri-pre-processing-container
  • 数据预处理

    • 从ADNI下载数据集

      https://adni.loni.usc.edu
    • 原始图像

      image-20240121120921751
    • 颅骨剥离

      image-20240119015518284
    • 配准(182x218x182)

    image-20240119015700542
    • 高斯平滑
    image-20240119015751695
    • 灰度归一化(使用fsleyes展示,itksnap只支持16位精度)

      image-20240121121120114
    • 切片(182x218x182)

      90,100,90

      9,18,9

      170,200,170

    • 切片

      • x轴切片 AD

        image-20240121121337456

        CN

        image-20240121121411338

        MCI

        image-20240121121512753

      • y轴切片

        AD

        image-20240121121842813

        CN

        image-20240121121903704 MCI

        image-20240121121926252

      • z轴切片 AD

        image-20240121122003221 CN

        image-20240121122018800 MCI

        image-20240121122040356

    • RESnet18(Pretrained=True)

About

This is gradesign

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published