SynFundus-1M合成眼底图像数据集
包含超过一百万张合成眼底图像和十五种标注类型的高质量数据集，用于眼底疾病诊断和模型预训练。
基本信息
资源简介
SynFundus-1M合成眼底图像数据集
【数据集背景】
本数据集源站标题为《parap1uie-s/SynFundus-1M》。源站首次发布：2023-12-13。
【数据内容】
源站原始描述：SynFundus-1M is a synthetic fundus image dataset consisting of millions of images and extensive annotations that can help improve the performance of diagnostic models. # SynFundus-1M SynFundus-1M is a synthetic fundus image dataset consisting of millions of images and extensive annotations that can help improve the performance of diagnostic models. Usage and License Notices: The data is intended and licensed for research use only. # News We have uploaded the dataset to huggingface dataset. Resea…
【适用方向】
影像类数据可用于医学图像分析模型（检测、分割、分类）的训练与评测，也可用于标注质量与成像参数相关性的研究。
【获取与许可】
源站页面：https://github.com/parap1uie-s/SynFundus-1M
使用许可：NOASSERTION。
【使用注意】
· 源站最后修改于 2024-07-11，引用时请注意版本时效。
下载信息
注册下载
需要注册 Kaggle 账号并登录后下载，适合需要跟踪下载记录和使用 API 的用户。
暂未开放公开下载
无需注册即可直接获取公开样本或文档，适合快速预览和评估数据集质量。
免登录当前数据集主要来源为 Kaggle 公开托管，完整影像包建议通过原始链接或 Kaggle API 获取。
使用方式
数据集获取
git clone https://github.com/parap1uie-s/SynFundus-1M.git
curl -L -o repo.zip https://github.com/parap1uie-s/SynFundus-1M/archive/refs/heads/main.zip
unzip repo.zip
源站 README 摘录（使用方式）
SynFundus-1M
SynFundus-1M is a synthetic fundus image dataset consisting of millions of images and extensive annotations that can help improve the performance of diagnostic models.
Usage and License Notices: The data is intended and licensed for research use only.
News
We have uploaded the dataset to huggingface dataset.
Researchers who cannot access BaiduNetDisk please provide huggingface username (re-fill in the application form).
Thanks for the help from Alen Li!
Demo visualize
SynFundus-1M images are annotated with 11 diseases types and 4 types of quality score.
Request SynFundus-1M
SynFundus-1M is for research purposes only, and researchers must comply with the CC-BY-NC-SA 4.0 agreement.
Researchers may send request via the application form.
Citation
If you find SynFundus-1M useful for your research and applications, please cite using this BibTeX:
@misc{shang2023synfundus,
title={SynFundus: A synthetic fundus images dataset with millions of samples and multi-disease annotations},
author={Fangxin Shang and Jie Fu and Yehui Yang and Haifeng Huang and Junwei Liu and Lei Ma},
year={2023},
eprint={2312.00377},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
License
The dataset is CC-BY-NC-SA 4.0 (allowing only non-commercial use) and models trained using the dataset should not be used outside of research purposes.
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。
需要用于商业化训练或医疗器械注册（NMPA/FDA）？
公开数据往往缺乏完整的病理回证与专家标注。千方医数集依托陕川双基地与超十家三甲医院直连，提供完全合规、去标识化、支持注册级质控的专病数据集交易服务。
