Objective:To observe the diagnostic value of six classification intelligent auxiliary diagnosis lightweight model for common fundus diseases based on fundus color photography.Methods:A applied research. A dataset of 2 400 color fundus images from Nanjing Medical University Eye Hospital and Zhejiang Mathematical Medical Society Smart Eye Database was collected, which was desensitized and labeled by a fundus specialist. Of these, 400 each were for diabetic retinopathy, glaucoma, retinal vein occlusion, high myopia, age-related macular degeneration, and normal fundus. The parameters obtained from the classical classification models VGGNet16, ResNet50, DenseNet121 and lightweight classification models MobileNet3, ShuffleNet2, GhostNet trained on the ImageNet dataset were migrated to the six-classified common fundus disease intelligent aid diagnostic model using a migration learning approach during training as initialization parameters for training to obtain the latest model. 1 315 color fundus images of clinical patients were used as the test set. Evaluation metrics included sensitivity, specificity, accuracy, F1-Score and agreement of diagnostic tests (Kappa value); comparison of subject working characteristic curves as well as area under the curve values for different models.Result:Compared with the classical classification model, the storage size and number of parameters of the three lightweight classification models were significantly reduced, with ShuffleNetV2 having an average recognition time per sheet 438.08 ms faster than the classical classification model VGGNet16. All 3 lightweight classification models had Accuracy > 80.0%; Kappa values > 70.0% with significant agreement; sensitivity, specificity, and F1-Score for the diagnosis of normal fundus images were ≥ 98.0%; Macro-F1 was 78.2%, 79.4%, and 81.5%, respectively.Conclusion:The intelligent assisted diagnosis of common fundus diseases based on fundus color photography is a lightweight model with high recognition accuracy and speed; the storage size and number of parameters are significantly reduced compared with the classical classification model.

Objective To propose a new evaluation system and evaluate the application value of diabetic retinopathy ( DR) intelligence assisted diagnostic technology based on fundus photography. Methods By using the diagnostic test method,an evaluation system of DR intelligent diagnostic technology based on fundus photography was established. The fundus photographs of 331 diabetic patients (662 eyes) with DR screening were collected in the First Affiliated Hospital of Huzhou University from January 2017 to October 2018. The results of experts ' diagnosis and intelligence assisted diagnosis were compared and evaluated. The evaluation system includes primary evaluation, intermediate evaluation and advanced evaluation. The primary evaluation is the consistency of non-DR ( NDR) in all diabetic patients receiving DR-assisted diagnostic techniques;the intermediate evaluation is the diagnosis consistency of DR lesion degree in patients diagnosed with DR (grade 1-4);the advanced evaluation is the diagnosis consistency of DR classification ( grade 0 -4 ) in all diabetic patients receiving DR-assisted diagnostic techniques. The intermediate evaluation includes two evaluation methods. The main evaluation indicators include sensitivity,specificity and Kappa value. Results Based on experts ' diagnosis, NDR accounted for 22. 7％;mild non-proliferative DR (NPDR),moderate NPDR,and severe NPDR accounted for 19. 9％,18. 7％ and 25. 7％,respectively;proliferative DR( PDR) accounted for 13. 0％. Based on intelligence diagnostic system,NDR accounted for 25. 8％;mild NPDR, moderate NPDR and severe NPDR accounted for 19. 7％,19. 3％ and 22. 8％,respectively;proliferative DR( PDR) accounted for 12. 4％. Based on evaluation system in the paper,the sensitivity,specificity and Kappa value in primary evaluation were 91. 4％, 84. 7％ and 0. 72;the sensitivity, specificity and Kappa value in intermediate evaluation method one were 88. 4％,91. 1％ and 0. 79;the sensitivity, specificity and Kappa value in intermediate evaluation method two were 80. 5％,93. 3％ and 0. 75;the Kappa value in advanced evaluation was 0. 62. Conclusions The evaluation system can be applied to the evaluation of DR intelligent diagnostic technology,and the evaluation result can be used as the basis for the selection of DR intelligent diagnosis application scene.