Oxford Eye Hospital, Oxford University Hospitals NHS Foundation Trust, Oxford, UK.
Nuffield Laboratory of Ophthalmology, Nuffield Department of Clinical Neurosciences, University of Oxford, Oxford, UK.
Acta Ophthalmol. 2020 Sep;98(6):e715-e721. doi: 10.1111/aos.14353. Epub 2020 Jan 24.
Recent advances in deep learning have seen an increase in its application to automated image analysis in ophthalmology for conditions with a high prevalence. We wanted to identify whether deep learning could be used for the automated classification of optical coherence tomography (OCT) images from patients with Stargardt disease (STGD) using a smaller dataset than traditionally used.
Sixty participants with STGD and 33 participants with a normal retinal OCT were selected, and a single OCT scan containing the centre of the fovea was selected as the input data. Two approaches were used: Model 1 - a pretrained convolutional neural network (CNN); Model 2 - a new CNN architecture. Both models were evaluated on their accuracy, sensitivity, specificity and Jaccard similarity score (JSS).
About 102 OCT scans from participants with a normal retinal OCT and 647 OCT scans from participants with STGD were selected. The highest results were achieved when both models were implemented as a binary classifier: Model 1 - accuracy 99.6%, sensitivity 99.8%, specificity 98.0% and JSS 0.990; Model 2 - accuracy 97.9%, sensitivity 97.9%, specificity 98.0% and JSS 0.976.
The deep learning classification models used in this study were able to achieve high accuracy despite using a smaller dataset than traditionally used and are effective in differentiating between normal OCT scans and those from patients with STGD. This preliminary study provides promising results for the application of deep learning to classify OCT images from patients with inherited retinal diseases.
深度学习的最新进展使得其在具有高患病率的眼科疾病的自动化图像分析中的应用有所增加。我们希望确定深度学习是否可以用于使用比传统方法更小的数据集对斯特格德氏病（STGD）患者的光学相干断层扫描（OCT）图像进行自动分类。
选择了 60 名 STGD 患者和 33 名正常视网膜 OCT 患者，将包含黄斑中心的单个 OCT 扫描作为输入数据。使用了两种方法：模型 1 - 预先训练的卷积神经网络（CNN）；模型 2 - 新的 CNN 架构。两种模型均在准确性、敏感性、特异性和 Jaccard 相似性评分（JSS）方面进行了评估。
选择了大约 102 个来自正常视网膜 OCT 患者的 OCT 扫描和 647 个来自 STGD 患者的 OCT 扫描。当两种模型都作为二进制分类器实施时，得到了最高的结果：模型 1 - 准确性 99.6%，敏感性 99.8%，特异性 98.0%和 JSS 0.990；模型 2 - 准确性 97.9%，敏感性 97.9%，特异性 98.0%和 JSS 0.976。
尽管使用的数据集比传统方法小，但本研究中使用的深度学习分类模型仍能达到高精度，并且能够有效区分正常 OCT 扫描和 STGD 患者的 OCT 扫描。这项初步研究为将深度学习应用于分类遗传性视网膜疾病患者的 OCT 图像提供了有前景的结果。
