Using Slit-Lamp Images for Deep Learning-Based Identification of Bacterial and Fungal Keratitis: Model Development and Validation with Different Convolutional Neural Networks.
Using Slit-Lamp Images for Deep Learning-Based Identification of Bacterial and Fungal Keratitis: Model Development and Validation with Different Convolutional Neural Networks.
复制标题
DOI：
10.3390/diagnostics11071246
复制
发表时间：
2021-07-12
期刊：
影响因子：
--
通讯作者：
Hsiao CH
中科院分区：
文献类型：
--
作者：
Hung N;Shih AK;Lin C;Kuo MT;Hwang YS;Wu WC;Kuo CF;Kang EY;Hsiao CH
在这项研究中，我们的目的是开发一个深度学习模型，利用裂隙灯图像识别细菌性角膜炎(BK)和真菌性角膜炎(FK)。我们在2010年1月1日至2019年12月31日期间从台湾的两个医疗中心回顾收集了培养证实的微生物角膜炎患者的裂隙灯图像。我们构建了一个深度学习算法，该算法由切割角膜图像的分割模型和应用不同卷积神经网络(CNN)来区分FK和BK的分类模型组成。CNN包括DenseNet121、DenseNet161、DenseNet169、DenseNet201、EfficientNetB3、InceptionV3、ResNet101和ResNet50。对模型性能进行了评估，并将其表示为接收器工作特性曲线的曲线下面积(AUC)。利用梯度加权类激活映射技术绘制了模型的热图。通过使用580例患者的1330幅图像，深度学习算法获得了最高的平均准确率80.0%。使用不同的CNN，BK的诊断准确率为79.6%~95.9%，FK的诊断准确率为26.3%~65.8%。DenseNet161的CNN显示出最好的模型性能，BK和FK的AUC都为0.85。热图显示，该模型能够识别角膜浸润物。该模型显示出比之前报道的普通眼科医生和角膜专家的诊断性能更好的诊断准确性。
In this study, we aimed to develop a deep learning model for identifying bacterial keratitis (BK) and fungal keratitis (FK) by using slit-lamp images. We retrospectively collected slit-lamp images of patients with culture-proven microbial keratitis between 1 January 2010 and 31 December 2019 from two medical centers in Taiwan. We constructed a deep learning algorithm consisting of a segmentation model for cropping cornea images and a classification model that applies different convolutional neural networks (CNNs) to differentiate between FK and BK. The CNNs included DenseNet121, DenseNet161, DenseNet169, DenseNet201, EfficientNetB3, InceptionV3, ResNet101, and ResNet50. The model performance was evaluated and presented as the area under the curve (AUC) of the receiver operating characteristic curves. A gradient-weighted class activation mapping technique was used to plot the heat map of the model. By using 1330 images from 580 patients, the deep learning algorithm achieved the highest average accuracy of 80.0%. Using different CNNs, the diagnostic accuracy for BK ranged from 79.6% to 95.9%, and that for FK ranged from 26.3% to 65.8%. The CNN of DenseNet161 showed the best model performance, with an AUC of 0.85 for both BK and FK. The heat maps revealed that the model was able to identify the corneal infiltrations. The model showed a better diagnostic accuracy than the previously reported diagnostic performance of both general ophthalmologists and corneal specialists.
登录
查看更多内容
影响因子：
3
作者：
Bhuiyan, Alauddin;Wong, Tien Yin;Smith, R. Theodore
通讯作者：
Smith, R. Theodore
影响因子：
19.5
作者：
Russakovsky, Olga;Deng, Jia;Fei-Fei, Li
通讯作者：
Fei-Fei, Li
影响因子：
12.8
作者：
Xu, Yesheng;Kong, Ming;Yao, Yu-Feng
通讯作者：
Yao, Yu-Feng
影响因子：
3.7
作者：
Kim SJ;Cho KJ;Oh S
通讯作者：
Oh S
影响因子：
4.4
作者：
Dalmon, Cyril;Porco, Travis C.;Acharya, Nisha R.
通讯作者：
Acharya, Nisha R.