MMRDR

Multi-Modal Retinal images for in-depth DR analysis (MMRDR)Diabetic Retinopathy (DR) is a leading cause of preventable blindness worldwide, particularly among working-age adults. We present a multi-modal retinal images dataset, comprising 11,118 CFP images, 10,404 UWF images and 2,938 OCT images. CFP and UWF images are annotated with 5 DR severity grades and 7 DR-related lesions, while OCT images are annotated with 3 DME severity classes.用于深度糖尿病视网膜病变分析的多模态视网膜影像数据集（MMRDR） 糖尿病视网膜病变（Diabetic Retinopathy, DR）是全球范围内可预防性失明的首要致病因素，尤其对劳动年龄成年人的视觉健康构成严重威胁。本研究提出一款多模态视网膜影像数据集，包含11118张彩色眼底照相（Color Fundus Photography, CFP）影像、10404张超广角眼底成像（Ultra-Widefield Fundus Photography, UWF）影像以及2938张光学相干断层扫描（Optical Coherence Tomography, OCT）影像。其中彩色眼底照相与超广角眼底成像影像均标注有5级糖尿病视网膜病变严重程度分级与7种糖尿病视网膜病变相关病灶；光学相干断层扫描影像则标注有3级糖尿病黄斑水肿（Diabetic Macular Edema, DME）严重程度等级。