智能化糖尿病眼底病变诊断系统

Diabetic Retinopathy (DR), the most common one of diabetic eye diseases, can cause loss of vision or blindness. We propose an automatic diabetic retinopathy diagnostic system to help patients know about their retinal conditions. The images are taken through the phone application and then transmitted to a cloud server to be analyzed, including localization of optic disk and macular, vessel segmentation, detection of lesions, and grading of DR. We use a multi-scale line operator to improve accuracy in segmenting small-scale vessels, a binary mask and image restoration to reduce the effect of the existence of vessels on optic disk localization. After the analysis, the fundus images are then graded as normal, mild Non-Proliferative Diabetic Retinopathy (NPDR), moderate NPDR or severe NPDR. The grading process uses region segmentation to improve the efficiency. The final grading results are tested based on the fundus images provided by the hospitals. We evaluate our system through comparing our grading results with those graded by experts, which comes out with an overall accuracy of up to 85%.