Department of Gastroenterology and Hepatology, Kindai University Faculty of Medicine, 377-2 Ohno-higashi, Osaka-sayama, Osaka, 589-8511, Japan.
Department of Human Health Sciences, Graduate School of Medicine, Kyoto University, 53 Kawahara-cho, Shogoin, Sakyo-ku, Kyoto, 606-8507, Japan.
J Gastroenterol. 2022 Apr;57(4):309-321. doi: 10.1007/s00535-022-01849-9. Epub 2022 Feb 27.
Ultrasonography (US) is widely used for the diagnosis of liver tumors. However, the accuracy of the diagnosis largely depends on the visual perception of humans. Hence, we aimed to construct artificial intelligence (AI) models for the diagnosis of liver tumors in US.
We constructed three AI models based on still B-mode images: model-1 using 24,675 images, model-2 using 57,145 images, and model-3 using 70,950 images. A convolutional neural network was used to train the US images. The four-class liver tumor discrimination by AI, namely, cysts, hemangiomas, hepatocellular carcinoma, and metastatic tumors, was examined. The accuracy of the AI diagnosis was evaluated using tenfold cross-validation. The diagnostic performances of the AI models and human experts were also compared using an independent test cohort of video images.
The diagnostic accuracies of model-1, model-2, and model-3 in the four tumor types are 86.8%, 91.0%, and 91.1%, whereas those for malignant tumor are 91.3%, 94.3%, and 94.3%, respectively. In the independent comparison of the AIs and physicians, the percentages of correct diagnoses (accuracies) by the AIs are 80.0%, 81.8%, and 89.1% in model-1, model-2, and model-3, respectively. Meanwhile, the median percentages of correct diagnoses are 67.3% (range 63.6%-69.1%) and 47.3% (45.5%-47.3%) by human experts and non-experts, respectively.
The performance of the AI models surpassed that of human experts in the four-class discrimination and benign and malignant discrimination of liver tumors. Thus, the AI models can help prevent human errors in US diagnosis.
超声检查（US）广泛用于肝脏肿瘤的诊断。然而，诊断的准确性在很大程度上取决于人类的视觉感知。因此，我们旨在构建用于 US 诊断肝脏肿瘤的人工智能（AI）模型。
我们构建了三个基于静态 B 型图像的 AI 模型：模型 1 使用 24675 张图像，模型 2 使用 57145 张图像，模型 3 使用 70950 张图像。使用卷积神经网络对 US 图像进行训练。AI 对四类肝脏肿瘤（囊肿、血管瘤、肝细胞癌和转移性肿瘤）进行分类。使用十折交叉验证评估 AI 诊断的准确性。还使用视频图像的独立测试队列比较 AI 模型和人类专家的诊断性能。
在四种肿瘤类型中，模型 1、模型 2 和模型 3 的诊断准确率分别为 86.8%、91.0%和 91.1%，而恶性肿瘤的诊断准确率分别为 91.3%、94.3%和 94.3%。在 AI 和医师的独立比较中，模型 1、模型 2 和模型 3 的 AI 正确诊断率（准确率）分别为 80.0%、81.8%和 89.1%。而人类专家和非专家的正确诊断率中位数分别为 67.3%（范围 63.6%-69.1%）和 47.3%（45.5%-47.3%）。
在四类肝脏肿瘤和良恶性肿瘤的鉴别中，AI 模型的性能优于人类专家。因此，AI 模型可以帮助防止 US 诊断中的人为错误。
