Detection of Diabetic Retinopathy via Pixel Color Amplification Using EfficientNetV2

This study adopts a pixel color amplification to increase the characteristics of the fundus image to solve inconsistent images quality problem and adopts EfficientNetV2 architecture of deep convolutional neural network (CNN) to detect Diabetic Retinopathy (DR). The results show that the proposed method can achieve 0.9120/87.16% of quadratic weight kappa and accuracy score and proves the efficacy of the proposed approach in DR classification. Thus, our work can help DR patients to reduce the probability of having lifelong blindness.