基于深度学习的红外眼底摄影分析，用于白内障合并糖尿病视网膜病变的自动诊断。
Deep learning-based analysis of infrared fundus photography for automated diagnosis of diabetic retinopathy with cataracts.
机构信息
From the Department of Ophthalmology, Shanghai Eye Disease Prevention & Treatment Center, Shanghai, China (Xue, Zou); Department of Ophthalmology, Shanghai General Hospital; National Clinical Research Center for Eye Diseases; Shanghai Key Laboratory of Ocular Fundus Diseases; Shanghai Engineering Center for Visual Science and Photomedicine; Shanghai Engineering Research Center of Precise Diagnosis and Treatment of Eye Diseases, Shanghai, China (Zhang, Ma, Zou); School of Computer Science, Shanghai Key Laboratory of Intelligent Information Processing, Fudan University, Shanghai, China (Hou, Feng); Academy for Engineering and Technology, Fudan University, Shanghai, China (Xiao, Feng, Zhao).
出版信息
J Cataract Refract Surg. 2023 Oct 1;49(10):1043-1048. doi: 10.1097/j.jcrs.0000000000001269.
PURPOSE
To develop deep learning-based networks for the diagnosis of diabetic retinopathy (DR) with cataracts based on infrared fundus images.
SETTING
Shanghai General Hospital, Shanghai Eye Disease Prevention & Treatment Center, Shanghai, China.
DESIGN
Development and evaluation of an artificial intelligence (AI) diagnostic method.
METHODS
A total of 10 665 infrared fundus images from 4553 patients with diabetes were used to train and test the model. For image quality assessment, left and right eye classification, DR diagnosis and grading, and segmentation of 3 DR lesions, an end-to-end software using EfficientNet and UNet was developed. The accuracy and performance of the software in comparison to human experts was evaluated.
RESULTS
The model achieved an accuracy of 75.31% for left and right eye classification, 100% for DR grading and diagnosis tasks, and 73.67% for internal test set, with corresponding areas under the curve (AUCs) of 0.88, 1.00, and 0.89, respectively. For DR lesion segmentation, the AUCs of hemorrhagic, microangioma, and exudative lesions were 0.86, 0.66, and 0.84, respectively. In addition, a contrast test of human-machine film reading confirmed the software's high sensitivity (96.3%) and specificity (90.0%) and consistency with the manual film reading group (κ = 0.869, P < .001). This easily deployable software generated reports quickly and promoted efficient DR screening with cataracts in clinical and community settings.
CONCLUSIONS
AI-assisted software can perform automatic analysis of infrared fundus images and has substantial application value for the diagnosis of DR patients with cataracts.
目的
基于红外眼底图像开发用于诊断伴白内障糖尿病视网膜病变（DR）的深度学习网络。
设置
中国上海，上海眼病防治中心，上海总医院。
设计
人工智能（AI）诊断方法的开发和评估。
方法
使用来自 4553 名糖尿病患者的 10665 张红外眼底图像来训练和测试模型。为了进行图像质量评估、左右眼分类、DR 诊断和分级以及 3 种 DR 病变的分割，开发了一个使用 EfficientNet 和 UNet 的端到端软件。评估了该软件与人类专家相比的准确性和性能。
结果
该模型在左右眼分类方面的准确率为 75.31%，在 DR 分级和诊断任务方面的准确率为 100%，在内部测试集方面的准确率为 73.67%，相应的曲线下面积（AUC）分别为 0.88、1.00 和 0.89。对于 DR 病变分割，出血、微血管瘤和渗出性病变的 AUC 分别为 0.86、0.66 和 0.84。此外，人机读片对比测试证实了该软件具有较高的灵敏度（96.3%）和特异性（90.0%），与手动读片组具有一致性（κ=0.869，P<0.001）。这款易于部署的软件可以快速生成报告，并促进临床和社区环境中伴白内障的 DR 高效筛查。
结论
AI 辅助软件可以对红外眼底图像进行自动分析，对伴白内障的 DR 患者的诊断具有重要的应用价值。
Zotero 插件