BenchReAD
收藏资源简介:
BenchReAD是一个全面的视网膜异常检测基准,包含两种最常用的视网膜成像模式:眼底摄影和光学相干断层扫描(OCT)。该基准包含大量数据集,涵盖了多种异常类型,能够进行严格的泛化评估。数据集由多个公开数据集整合而成,包括EDDFS、BRSET、RIADD、JSIEC、OCT 2017、OCTDL和OCTID等。BenchReAD旨在为视网膜异常检测提供一个系统性和全面性的基准,促进该领域的发展。
BenchReAD is a comprehensive benchmark for retinal abnormality detection, encompassing the two most widely utilized retinal imaging modalities: fundus photography and optical coherence tomography (OCT). It incorporates a large-scale collection of datasets spanning diverse abnormality categories, enabling rigorous generalization assessment. This benchmark is compiled from multiple publicly available datasets, such as EDDFS, BRSET, RIADD, JSIEC, OCT 2017, OCTDL, OCTID, and others. BenchReAD is designed to offer a systematic and comprehensive benchmark for retinal abnormality detection, thereby facilitating the advancement of the relevant research field.
BenchReAD 数据集概述
数据集简介
- 官方代码库:BenchReAD: A systematic benchmark for retinal anomaly detection (MICCAI 2025)
- 包含眼底图像(fundus)和光学相干断层扫描(OCT)两个基准数据集
数据集构建
1. 眼底图像基准数据集
训练集与验证集
- 数据来源:
- EDDFS数据集(https://github.com/xia-xx-cv/EDDFS_dataset)
- BRSET数据集(https://physionet.org/content/brazilian-ophthalmological/1.0.1/)
- 预处理步骤:
- EDDFS数据预处理使用EDDFS_preprocessing.ipynb
- EDDFS数据分割使用EDDFS_split.ipynb
- BRSET数据预处理使用BRSET_preprocessing.ipynb
- BRSET数据分割使用BRSET_split.ipynb
测试集
- 数据来源:
- RIADD数据集(https://riadd.grand-challenge.org/)
- JSIEC数据集(https://www.kaggle.com/datasets/linchundan/fundusimage1000)
- 预处理步骤:
- RIADD数据预处理使用RIADD_preprocessing.ipynb
- RIADD数据分割使用RIADD_split.ipynb
- JSIEC数据预处理使用JSIEC.ipynb
2. OCT基准数据集
训练集与验证集
- 数据来源:
- OCT 2017数据集(https://www.kaggle.com/datasets/paultimothymooney/kermany2018)
- 预处理步骤:
- OCT 2017数据预处理使用OCT_2017_preprocessing.ipynb
- OCT 2017数据分割使用OCT_2017_split.ipynb
测试集
- 数据来源:
- OCTDL数据集(https://data.mendeley.com/datasets/sncdhf53xc/4)
- OCTID数据集(https://www.kaggle.com/datasets/saifurrahmanshatil/retinal-oct-dataset)
- 预处理步骤:
- OCTDL数据预处理使用OCTDL.ipynb
- OCTID数据预处理使用OCTID_step1.ipynb和OCTID_step2.ipynb
参考文献
- Xia, X., et al. Benchmarking deep models on retinal fundus disease diagnosis and a large-scale dataset. Signal Processing: Image Communication 127, 117151 (2024).
- Nakayama, L.F., et al. Brset: A brazilian multilabel ophthalmological dataset of retina fundus photos. PLOS Digital Health 3(7), e0000454 (2024).
- Pachade, S., et al. Retinal fundus multi-disease image dataset (rfmid): a dataset for multi-disease detection research. Data 6(2), 14 (2021)
- Cen, L.P., et al. Automatic detection of 39 fundus diseases and conditions in retinal photographs using deep neural networks. Nature communications 12(1), 4828 (2021)
- Kermany, D.S., et al. Identifying medical diagnoses and treatable diseases by image-based deep learning. cell 172(5), 1122–1131 (2018)
- Kulyabin, M., et al. Octdl: Optical coherence tomography dataset for image-based deep learning methods. Scientific Data 11(1), 365 (2024)
- Gholami, P., et al. Octid: Optical coherence tomography image database. Computers & Electrical Engineering 81, 106532 (2020)




