数据集获取

git clone https://github.com/vafaei-ar/medical-datasets.git

curl -L -o repo.zip https://github.com/vafaei-ar/medical-datasets/archive/refs/heads/master.zip
unzip repo.zip

源站 README 摘录（使用方式）

medical-datasets

I list all the medical data sets I find in my researches.

IMAGES

• Zhao et al. have released a COVID19-CT dataset, with detailed descriptions.
  You can find it here.
• Prostate cANcer graDe Assessment (PANDA) is a challenge on the Kaggle, so you can find the data and description here
• There are 3 challenges for aneurysm detection, segmentation and rupture risk estimation. You can find the challenge here. Registration may be needed.
• The AutoImplant challenge where a fast and automatic design of cranial implants is highly desired. You can find the challenge information here. Registration may be needed.
• DFU challenge is a repository of 4500 DFU images for the purpose of supporting research toward more advanced methods of Diabetic Foot Ulcers (DFU) detection. You can access the data here. Registration may be needed.
• KNee OsteoArthritis Prediction (KNOAP2020) Challenge is about identifying which knees will develop symptomatic radiographic knee osteoarthritis within 6.5 years follow-up. More information is provided here. Registration may be needed.
• The Learn2reg challenge, which builds on a popular tutorial in 2019, will be a simplified challenge design that removes many of the common pitfalls for learning and applying transformations. The data can either be individually or comprehensively addressed by participants and cover both intra- and inter-patient alignment, CT, ultrasound and MRI modalities, neuro-, thorax and abdominal anatomies and the four of the imminent challenges of medical image registration:
  1. learning from small datasets
  2. estimating large deformations
  3. dealing with multi-modal scans
  4. learning from noisy annotations
     you can find more information here.
• The SIIM-ACR Pneumothorax segmentation challenge is about developing an AI algorithm to detect pneumothorax. More info is provided here.
• The goal of APTOS 2019 Blindness Detection challenge is to gain the ability to automatically

数据加载示例（图像类）

from PIL import Image
import glob, os

files = (glob.glob(os.path.join(path, "**", "*.png"), recursive=True)
       + glob.glob(os.path.join(path, "**", "*.jpg"), recursive=True)
       + glob.glob(os.path.join(path, "**", "*.tif"), recursive=True))
print("图像文件数:", len(files))
img = Image.open(files[0]); print("尺寸/模式:", img.size, img.mode)

# torchvision Dataset 方式:
# from torchvision import datasets
# ds = datasets.ImageFolder(path)  # 要求 子目录=类别

目录组织与标注格式以源站说明和下载后实际文件为准。

完整仓库：github.com/vafaei-ar/medical-datasets