应用人工智能识别超广角眼底照相多病种的初步研究

Preliminary study on the application of artificial intelligence to identify multiple diseases in ultra-widefield fundus images

目的构建一个小样本超广角眼底照相(UWFI)多疾病分类人工智能模型,初步探究人工智能对UWFI多病种分类任务的能力。方法回顾性研究。2016年至2021年于武汉大学人民医院眼科就诊并行UWFI检查的1123例患者的1608张图像用于UWFI多疾病分类人工智能模型构建。其中,糖尿病视网膜病变(DR)、视网膜静脉阻塞(RVO)、病理性近视(PM)、视网膜脱离(RD)、正常眼底图像分别为320、330、319、268、371张。来自天津医科大学眼科医院106例患者的135张图像作为外部测试集。选取EfficientNet-B7作为主干网络,对纳入的UWFI图像进行分类分析。使用受试者工作特征曲线及曲线下面积(AUC)、灵敏度、特异性、准确率评估分类模型在测试集上的表现,所有数据均使用数值及95%可信区间(CI)表达。将数据集在网络模型ResNet50、ResNet101上进行训练,并在外部测试集上进行测试,对比观察EfficientNet与上述两种模型的性能。结果UWFI多疾病分类人工智能模型在内部、外部测试集上的总分类准确率分别为92.57%(95%CI 91.13%~92.92%)、88.89%(95%CI 88.11%~90.02%)。其中,正常眼底分别为96.62%、92.59%,DR分别为95.95%、95.56%,RVO分别为96.62%、98.52%,PM分别为98.65%、97.04%,RD分别为97.30%、94.07%。在内部、外部测试集上的平均AUC分别为0.993、0.983。其中,正常眼底分别为0.994、0.939,DR分别为0.999、0.995,RVO分别为0.985、1.000,PM分别为0.991、0.993,RD分别为0.995、0.990。内部、外部测试集上EfficientNet性能均较ResNet50、ResNet101模型更佳。结论初步构建的小样本UWFI多疾病分类人工智能模型对常见眼底疾病的分类水平较高,可能具有辅助临床筛查及诊断的价值。

Objective To build a small-sample ultra-widefield fundus images(UWFI)multi-disease classification artificial intelligence model,and initially explore the ability of artificial intelligence to classify UWFI multi-disease tasks.Methods A retrospective study.From 2016 to 2021,1608 images from 1123 patients who attended the Eye Center of the Renmin Hospital of Wuhan University and underwent UWFI examination were used for UWFI multi-disease classification artificial intelligence model construction.Among them,320,330,319,268,and 371 images were used for diabetic retinopathy(DR),retinal vein occlusion(RVO),pathological myopia(PM),retinal detachment(RD),and normal fundus images,respectively.135 images from 106 patients at the Tianjin Medical University Eye Hospital were used as the external test set.EfficientNet-B7 was selected as the backbone network for classification analysis of the included UWFI images.The performance of the UWFI multi-task classification model was assessed using the receiver operating characteristic curve,area under the curve(AUC),sensitivity,specificity,and accuracy.All data were expressed using numerical values and 95%confidence intervals(CI).The datasets were trained on the network models ResNet50 and ResNet101 and tested on an external test set to compare and observe the performance of EfficientNet with the 2 models mentioned above.Results The overall classification accuracy of the UWFI multi-disease classification artificial intelligence model on the internal and external test sets was 92.57%(95%CI91.13%-92.92%)and 88.89%(95%CI88.11%-90.02%),respectively.These were 96.62%and 92.59%for normal fundus,95.95%and 95.56%for DR,96.62%and 98.52%for RVO,98.65%and 97.04%for PM,and 97.30%and 94.07%for RD,respectively.The mean AUC on the internal and external test sets was 0.993 and 0.983,respectively,with 0.994 and 0.939 for normal fundus,0.999 and 0.995 for DR,0.985 and 1.000 for RVO,0.991 and 0.993 for PM and 0.995 and 0.990 for RD,respectively.EfficientNet performed better than the ResNet50 and ResNetlOl models on both the internal and external test sets.Conclusion The preliminary UWFI multi-disease classification artificial intelligence model using small samples constructed in this study is able to achieve a high accuracy rate,and the model may have some value in assisting clinical screening and diagnosis.