Wan Cheng, Fang Jiyi, Hua Xiao, Chen Lu, Zhang Shaochong, Yang Weihua
College of Electronic Information Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing, China.
Nanjing Star-mile Technology Co., Ltd., Nanjing, China.
Front Comput Neurosci. 2023 Apr 20;17:1169464. doi: 10.3389/fncom.2023.1169464. eCollection 2023.
To propose a five-category model for the automatic detection of myopic macular lesions to help grassroots medical institutions conduct preliminary screening of myopic macular lesions from limited number of color fundus images.
First, 1,750 fundus images of non-myopic retinal lesions and four categories of pathological myopic maculopathy were collected, graded, and labeled. Subsequently, three five-classification models based on Vision Outlooker for Visual Recognition (VOLO), EfficientNetV2, and ResNet50 for detecting myopic maculopathy were trained with data-augmented images, and the diagnostic results of the different trained models were compared and analyzed. The main evaluation metrics were sensitivity, specificity, negative predictive value (NPV), positive predictive value (PPV), area under the curve (AUC), kappa and accuracy, and receiver operating characteristic curve (ROC).
The diagnostic accuracy of the VOLO-D2 model was 96.60% with a kappa value of 95.60%. All indicators used for the diagnosis of myopia-free macular degeneration were 100%. The sensitivity, NPV, specificity, and PPV for diagnosis of leopard fundus were 96.43, 98.33, 100, and 100%, respectively. The sensitivity, specificity, PPV, and NPV for the diagnosis of diffuse chorioretinal atrophy were 96.88, 98.59, 93.94, and 99.29%, respectively. The sensitivity, specificity, PPV, and NPV for the diagnosis of patchy chorioretinal atrophy were 92.31, 99.26, 97.30, and 97.81%, respectively. The sensitivity, specificity, PPV, and NPV for the diagnosis of macular atrophy were 100, 98.10, 84.21, and 100%, respectively.
The VOLO-D2 model accurately identified myopia-free macular lesions and four pathological myopia-related macular lesions with high sensitivity and specificity. It can be used in screening pathological myopic macular lesions and can help ophthalmologists and primary medical institution providers complete the initial screening diagnosis of patients.
提出一种用于自动检测近视性黄斑病变的五类模型，以帮助基层医疗机构从有限数量的彩色眼底图像中对近视性黄斑病变进行初步筛查。
首先，收集1750张非近视性视网膜病变和四类病理性近视性黄斑病变的眼底图像，进行分级和标注。随后，使用数据增强图像对基于视觉识别的视觉展望者（VOLO）、高效神经网络V2（EfficientNetV2）和残差网络50（ResNet50）的三种五类模型进行训练，以检测近视性黄斑病变，并对不同训练模型的诊断结果进行比较和分析。主要评估指标为敏感性、特异性、阴性预测值（NPV）、阳性预测值（PPV）、曲线下面积（AUC）、kappa值和准确性，以及受试者工作特征曲线（ROC）。
VOLO-D2模型的诊断准确率为96.60%，kappa值为95.60%。用于诊断无近视性黄斑变性的所有指标均为100%。诊断豹纹状眼底的敏感性、NPV、特异性和PPV分别为96.43%、98.33%、100%和100%。诊断弥漫性脉络膜视网膜萎缩的敏感性、特异性、PPV和NPV分别为96.88%、98.59%、93.94%和99.29%。诊断斑片状脉络膜视网膜萎缩的敏感性、特异性、PPV和NPV分别为92.31%、99.26%、97.30%和97.81%。诊断黄斑萎缩的敏感性、特异性、PPV和NPV分别为100%、98.10%、84.21%和100%。
VOLO-D2模型以高敏感性和特异性准确识别了无近视性黄斑病变和四种与病理性近视相关的黄斑病变。它可用于筛查病理性近视性黄斑病变，并可帮助眼科医生和基层医疗机构提供者完成患者的初步筛查诊断。
