Background: With the diabetes mellitus (DM) prevalence increasing annually, the human grading of retinal images\nto evaluate DR has posed a substantial burden worldwide. SmartEye is a recently developed fundus image\nprocessing and analysis system with lesion quantification function for DR screening. It is sensitive to the lesion area\nand can automatically identify the lesion position and size. We reported the diabetic retinopathy (DR) grading\nresults of SmartEye versus ophthalmologists in analyzing images captured with non-mydriatic fundus cameras in\ncommunity healthcare centers, as well as DR lesion quantitative analysis results on different disease stages.\nMethods: This is a cross-sectional study. All the fundus images were collected from the Shanghai Diabetic Eye\nStudy in Diabetics (SDES) program from Apr 2016 to Aug 2017. 19,904 fundus images were acquired from 6013\ndiabetic patients. The grading results of ophthalmologists and SmartEye are compared. Lesion quantification of\nseveral images at different DR stages is also presented.\nResults: The sensitivity for diagnosing no DR, mild NPDR (non-proliferative diabetic retinopathy), moderate NPDR,\nsevere NPDR, PDR (proliferative diabetic retinopathy) are 86.19, 83.18, 88.64, 89.59, and 85.02%. The specificity are\n63.07, 70.96, 64.16, 70.38, and 74.79%, respectively. The AUC are PDR, 0.80 (0.79, 0.81); severe NPDR, 0.80 (0.79, 0.80);\nmoderate NPDR, 0.77 (0.76, 0.77); and mild NPDR, 0.78 (0.77, 0.79). Lesion quantification results showed that the\ntotal hemorrhage area, maximum hemorrhage area, total exudation area, and maximum exudation area increase\nwith DR severity.\nConclusions: SmartEye has a high diagnostic accuracy in DR screening program using non-mydriatic fundus\ncameras. SmartEye quantitative analysis may be an innovative and promising method of DR diagnosis and grading.
Loading....