Department of Molecular Pharmacology, National Institute of Neuroscience, National Center of Neurology and Psychiatry, Kodaira, Japan; Department of Life Science and Medical Bioscience, Graduate School of Advanced Science and Engineering, Waseda University, Shinjuku, Tokyo, Japan.
Department of Systems Life Engineering, Maebashi Institute of Technology, Maebashi, Gunma, Japan.
Comput Biol Med. 2021 Oct;137:104795. doi: 10.1016/j.compbiomed.2021.104795. Epub 2021 Aug 25.
Diabetic retinopathy (DR) has become one of the major causes of blindness. Due to the increased prevalence of diabetes worldwide, diabetic patients exhibit high probabilities of developing DR. There is a need to develop a labor-less computer-aided diagnosis system to support the clinical diagnosis. Here, we attempted to develop simple methods for severity grading and lesion detection from retinal fundus images. We developed a severity grading system for DR by transfer learning with a recent convolutional neural network called EfficientNet-B3 and the publicly available Kaggle Asia Pacific Tele-Ophthalmology Society (APTOS) 2019 training dataset, which includes artificial noise. After removing the blurred and duplicated images from the dataset using a numerical threshold, the trained model achieved specificity and sensitivity values ≳ 0.98 in the identification of DR retinas. For severity grading, the classification accuracy values of 0.84, 0.95, and 0.98 were recorded for the 1st, 2nd, and 3rd predicted labels, respectively. The utility of EfficientNets-B3 for the severity grading of DR as well as the detailed retinal areas referred were confirmed via visual explanation methods of convolutional neural networks. Lesion extraction was performed by applying an empirically defined threshold value to the enhanced retinal images. Although the extraction of blood vessels and detection of red lesions occurred simultaneously, the red and white lesions, including both soft and hard exudates, were clearly extracted. The detected lesion areas were further confirmed with ground truth using the DIARETDB1 database images with general accuracy. The simple and easily applicable methods proposed in this study will aid in the detection and severity grading of DR, which might help in the selection of appropriate treatment strategies for DR.
糖尿病视网膜病变（DR）已成为主要致盲原因之一。由于全球糖尿病患病率的增加，糖尿病患者发展为 DR 的概率很高。因此，需要开发一种无需人工的计算机辅助诊断系统来支持临床诊断。在这里，我们尝试从眼底图像中开发简单的严重程度分级和病变检测方法。我们通过使用最近的卷积神经网络 EfficientNet-B3 和公开的 Kaggle 亚太远程眼科学会（APTOS）2019 训练数据集（其中包括人工噪声）进行迁移学习，开发了一种 DR 严重程度分级系统。从数据集中使用数值阈值去除模糊和重复的图像后，经过训练的模型在识别 DR 视网膜方面特异性和敏感性值≳0.98。对于严重程度分级，记录到 1 级、2 级和 3 级预测标签的分类准确率分别为 0.84、0.95 和 0.98。通过卷积神经网络的可视化解释方法，确认了 EfficientNets-B3 对 DR 严重程度分级的有效性以及所涉及的详细视网膜区域。通过对增强的视网膜图像应用经验定义的阈值来进行病变提取。尽管血管和红色病变的提取同时进行，但红色和白色病变，包括软性和硬性渗出物，都被清晰地提取出来。使用一般准确性的 DIARETDB1 数据库图像与地面实况进一步确认检测到的病变区域。本研究提出的简单且易于应用的方法将有助于 DR 的检测和严重程度分级，这可能有助于为 DR 选择适当的治疗策略。
