Division of Gastroenterology and Hepatology, Department of Internal Medicine, Taipei Medical University Hospital, Taipei, Taiwan, ROC.
Division of Gastroenterology and Hepatology, Department of Internal Medicine, School of Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan, ROC.
J Chin Med Assoc. 2021 Sep 1;84(9):842-850. doi: 10.1097/JCMA.0000000000000585.
The prevalence of nonalcoholic fatty liver disease is increasing over time worldwide, with similar trends to those of diabetes and obesity. A liver biopsy, the gold standard of diagnosis, is not favored due to its invasiveness. Meanwhile, noninvasive evaluation methods of fatty liver are still either very expensive or demonstrate poor diagnostic performances, thus, limiting their applications. We developed neural network-based models to assess fatty liver and classify the severity using B-mode ultrasound (US) images.
We followed standards for reporting of diagnostic accuracy guidelines to report this study. In this retrospective study, we utilized B-mode US images from a consecutive series of patients to develop four-class, two-class, and three-class diagnostic prediction models. The images were eligible if confirmed by at least two gastroenterologists. We compared pretrained convolutional neural network models, consisting of visual geometry group (VGG)19, ResNet-50 v2, MobileNet v2, Xception, and Inception v2. For validation, we utilized 20% of the dataset resulting in >100 images for each severity category.
There were 21,855 images from 2,070 patients classified as normal (N = 11,307), mild (N = 4,467), moderate (N = 3,155), or severe steatosis (N = 2,926). We used ResNet-50 v2 for the final model as the best ones. The areas under the receiver operating characteristic curves were 0.974 (mild steatosis vs others), 0.971 (moderate steatosis vs others), 0.981 (severe steatosis vs others), 0.985 (any severity vs normal), and 0.996 (moderate-to-severe steatosis/clinically abnormal vs normal-to-mild steatosis/clinically normal).
Our deep learning models achieved comparable predictive performances to the most accurate, yet expensive, noninvasive diagnostic methods for fatty liver. Because of the discriminative ability, including for mild steatosis, significant impacts on clinical applications for fatty liver are expected. However, we need to overcome machine-dependent variation, motion artifacts, lacking of second confirmation from any other tools, and hospital-dependent regional bias.
非酒精性脂肪性肝病的患病率在全球范围内呈上升趋势，与糖尿病和肥胖的趋势相似。肝活检是诊断的金标准，但由于其侵袭性而不受青睐。同时，脂肪性肝病的非侵入性评估方法仍然非常昂贵或诊断性能不佳，因此限制了它们的应用。我们开发了基于神经网络的模型，使用 B 型超声（US）图像评估脂肪肝并对其严重程度进行分类。
我们遵循诊断准确性报告标准指南来报告本研究。在这项回顾性研究中，我们利用连续系列患者的 B 型 US 图像来开发四分类、二分类和三分类诊断预测模型。如果至少有两名胃肠病学家确认，这些图像才有资格入选。我们比较了预训练的卷积神经网络模型，包括视觉几何组（VGG）19、ResNet-50 v2、MobileNet v2、Xception 和 Inception v2。为了验证，我们利用了数据集的 20%，结果每个严重程度类别有超过 100 张图像。
共有 2070 名患者的 21855 张图像被分类为正常（N=11307）、轻度（N=4467）、中度（N=3155）或重度脂肪变性（N=2926）。我们最终选择了 ResNet-50 v2 作为最佳模型。受试者工作特征曲线下的面积分别为 0.974（轻度脂肪变性与其他）、0.971（中度脂肪变性与其他）、0.981（重度脂肪变性与其他）、0.985（任何严重程度与正常）和 0.996（中重度脂肪变性/临床异常与正常至轻度脂肪变性/临床正常）。
我们的深度学习模型在预测性能上与最准确但昂贵的脂肪性肝病非侵入性诊断方法相当。由于其鉴别能力，包括对轻度脂肪变性的鉴别能力，预计对脂肪性肝病的临床应用会产生重大影响。然而，我们需要克服机器依赖性的变化、运动伪影、缺乏来自任何其他工具的第二次确认以及医院依赖性的区域偏差。
