Automated detection of myopic maculopathy from color fundus photographs using deep convolutional neural networks  

作　　者：Jun Li Lilong Wang Yan Gao Qianqian Liang Lingzhi Chen Xiaolei Sun Huaqiang Yang Zhongfang Zhao Lina Meng Shuyue Xue Qing Du zhichun Zhang Chuanfeng LV Haifeng Xu Zhen Guo Guotong Xie Lixin Xie 

机构地区：[1]Qingdao Eye Hospital of Shandong First Medical University,5 Yanerdao Road,Qingdao 266071,China [2]State Key Laboratory Cultivation Base,Shandong Provincial Key Laboratory of Ophthalmology,Shandong Eye Institute,Shandong First Medical University&Shandong Academy of Medical Sciences,Qingdao 266071,China [3]Ping An Healthcare Technology,9F Building B,PingAn IFC,No.1-3 Xinyuan South Road,Bejing 100027,China [4]Shandong Eye Hospital of Shandong First Medical University,Jinan 250021,China [5]Rongcheng Eye Hospital,Weihai 264200,China [6]Hospital of PLA Navy,Qingdao 266071,China [7]Qilu Hospital of Shandong University(Qingdao),Qingdao 266035,China [8]Ping An Healthcare and Technology Company Limited,hanghai 200030,China [9]ping An International Smart City Technology Company Limited,Shenzhen 518000,China.

出　　处：《Eye and Vision》2024年第4期11-22,共12页眼视光学杂志（英文）

基　　金：The research has been supported by the Qingdao Science and Technology Demonstration and Guidance Project(Grant No.20-3-4-45-nsh);Academic Promotion Plan of Shandong First Medical University&Shandong Academy of Medical Sciences(Grant No.2019ZL001);National Science and Technology Major Project of China(Grant No.2017ZX09304010).

摘　　要：Background:Myopic maculopathy(MM)has become a major cause of visual impairment and blindness worldwide,especially in East Asian countries.Deep learning approaches such as deep convolutional neural networks(DCNN)have been successfully applied to identify some common retinal diseases and show great potential for the intelligent analysis of MM.This study aimed to build a reliable approach for automated detection of MM from retinal fundus images using DCNN models.Methods:A dual-stream DCNN(DCNN-DS)model that perceives features from both original images and corresponding processed images by color histogram distribution optimization method was designed for classification of no MM,tessellated fundus(TF),and pathologic myopia(PM).A total of 36,515 gradable images from four hospitals were used for DCNN model development,and 14,986 gradable images from the other two hospitals for external testing.We also compared the performance of the DCNN-DS model and four ophthalmologists on 3000 randomly sampledfundus images.Results:The DCNN-DS model achieved sensitivities of 93.3%and 91.0%,specificities of 99.6%and 98.7%,areas under the receiver operating characteristic curves(AUCs)of 0.998 and 0.994 for detecting PM,whereas sensitivities of 98.8%and 92.8%,specificities of 95.6%and 94.1%,AUCs of 0.986 and 0.970 for detecting TF in two external testing datasets.In the sampled testing dataset,the sensitivities of four ophthalmologists ranged from 88.3%to 95.8%and 81.1%to 89.1%,and the specificities ranged from 95.9%to 99.2%and 77.8%to 97.3%for detecting PM and TF,respectively.Meanwhile,the DCNN-DS model achieved sensitivities of 90.8%and 97.9%and specificities of 99.1%and 94.0%for detecting PMand T,respectively.Conclusions:The proposed DCNN-DS approach demonstrated reliable performance with high sensitivity,specificity,and AUC to classify different MM levels on fundus photographs sourced from clinics.It can help identify MM automatically among the large myopic groups and show great potential for real-life applications.

关 键 词：Myopic maculopathy Tessellated fundus Pathologic myopia Deep convolutional neural network Color fundus image 

分 类 号：R774.5[医药卫生—眼科]