Synthetic Retinal OCT Dataset Generated using Class-Conditioned DDPM

This dataset contains synthetic Optical Coherence Tomography (OCT) retinal images generated using class conditioned denoising diffusion probabilistic models (DDPMs). The images represent four diagnostic categories: choroidal neovascularization (CNV), diabetic macular edema (DME), drusen, and normal retina. Synthetic data are generated to address the limited availability of labeled OCT datasets and support the development of AI models for retinal disease classification. The generated images aim to preserve realistic retinal structures and disease characteristics for machine learning research.本数据集包含利用类条件降噪扩散概率模型（class conditioned denoising diffusion probabilistic models，DDPMs）生成的合成光学相干断层扫描（Optical Coherence Tomography，OCT）视网膜图像。该数据集涵盖四大诊断类别：脉络膜新生血管（choroidal neovascularization，CNV）、糖尿病性黄斑水肿（diabetic macular edema，DME）、玻璃膜疣以及正常视网膜。构建此类合成数据旨在解决标注型OCT数据集稀缺的问题，并助力视网膜疾病分类AI模型的研发。所生成的图像旨在保留真实的视网膜结构与疾病特征，以支撑机器学习相关研究。