Intelligent diagnostic model for pterygium by combining attention mechanism and MobileNetV2  

作　　者：Mao-Nian Wu Kai He Yi-Bei Yu Bo Zheng Shao-Jun Zhu Xiang-Qian Hong Wen-Qun Xi Zhe Zhang 

机构地区：[1]School of Information Engineering,Huzhou University,Huzhou 313000,Zhejiang Province,China [2]Zhejiang Province Key Laboratory of Smart Management and Application of Modern Agricultural Resources,Huzhou University,Huzhou 313000,Zhejiang Province,China [3]School of Mathematical Information,Shaoxing University,Shaoxing 312000,Zhejiang Province,China [4]Shenzhen Eye Institute,Shenzhen Eye Hospital,Jinan University,Shenzhen 518040,Guangdong Province,China

出　　处：《International Journal of Ophthalmology(English edition)》2024年第7期1184-1192,共9页国际眼科杂志（英文版）

基　　金：Supported by the National Natural Science Foundation of China(No.61906066);Scientific Research Fund of Zhejiang Provincial Education Department(No.Y202147191);Huzhou University Graduate Research Innovation Project(No.2020KYCX21);Sanming Project of Medicine in Shenzhen(SZSM202311012);Shenzhen Science and Technology Program(No.JCYJ20220530153604010).

摘　　要：AIM:To evaluate the application of an intelligent diagnostic model for pterygium.METHODS:For intelligent diagnosis of pterygium,the attention mechanisms—SENet,ECANet,CBAM,and Self-Attention—were fused with the lightweight MobileNetV2 model structure to construct a tri-classification model.The study used 1220 images of three types of anterior ocular segments of the pterygium provided by the Eye Hospital of Nanjing Medical University.Conventional classification models—VGG16,ResNet50,MobileNetV2,and EfficientNetB7—were trained on the same dataset for comparison.To evaluate model performance in terms of accuracy,Kappa value,test time,sensitivity,specificity,the area under curve(AUC),and visual heat map,470 test images of the anterior segment of the pterygium were used.RESULTS:The accuracy of the MobileNetV2+Self-Attention model with 281 MB in model size was 92.77%,and the Kappa value of the model was 88.92%.The testing time using the model was 9ms/image in the server and 138ms/image in the local computer.The sensitivity,specificity,and AUC for the diagnosis of pterygium using normal anterior segment images were 99.47%,100%,and 100%,respectively;using anterior segment images in the observation period were 88.30%,95.32%,and 96.70%,respectively;and using the anterior segment images in the surgery period were 88.18%,94.44%,and 97.30%,respectively.CONCLUSION:The developed model is lightweight and can be used not only for detection but also for assessing the severity of pterygium.

关 键 词：deep learning attention mechanism PTERYGIUM intelligent diagnosis 

分 类 号：R77[医药卫生—眼科]