Purpose: Liver imaging reporting and data system (LI-RADS) classification, especially the identification of LR-3 to 5 lesions with hepatocellular carcinoma (HCC) probability, is of great significance to treatment strategy determination. We aimed to develop a semi-automatic LI-RADS grading system on multiphase gadoxetic acid-enhanced MRI using deep convolutional neural networks (CNN).
Patients and Methods: An internal data set of 439 patients and external data set of 71 patients with suspected HCC were included and underwent gadoxetic acid-enhanced MRI. The expert-guided LI-RADS grading system consisted of four deep 3D CNN models including a tumor segmentation model for automatic diameter estimation and three classification models of LI-RADS major features including arterial phase hyper-enhancement (APHE), washout and enhancing capsule. An end-to-end learning system comprising single deep CNN model that directly classified the LI-RADS grade was developed for comparison.
Results: On internal testing set, the segmentation model reached a mean dice of 0.84, with the accuracy of mapped diameter intervals as 82.7% (95% CI: 74.4%, 91.7%). The area under the curves (AUCs) were 0.941 (95% CI: 0.914, 0.961), 0.859 (95% CI: 0.823, 0.890) and 0.712 (95% CI: 0.668, 0.754) for APHE, washout and capsule, respectively. The expert-guided system significantly outperformed the end-to-end system with a LI-RADS grading accuracy of 68.3% (95% CI: 60.8%, 76.5%) vs 55.6% (95% CI: 48.8%, 63.0%) (P< 0.0001). On external testing set, the accuracy of mapped diameter intervals was 91.5% (95% CI: 81.9%, 100.0%). The AUCs were 0.792 (95% CI: 0.745, 0.833), 0.654 (95% CI: 0.602, 0.703) and 0.658 (95% CI: 0.606, 0.707) for APHE, washout and capsule, respectively. The expert-guided system achieved an overall grading accuracy of 66.2% (95% CI: 58.0%, 75.2%), significantly higher than the end-to-end system of 50.1% (95% CI: 43.1%, 58.1%) (P< 0.0001).
Conclusion: We developed a semi-automatic step-by-step expert-guided LI-RADS grading system (LR-3 to 5), superior to the conventional end-to-end learning system. This deep learning-based system may improve workflow efficiency for HCC diagnosis in clinical practice.

Keywords: liver imaging reporting and data system, LI-RADS, hepatocellular carcinoma, HCC, magnetic resonance imaging, MRI, deep learning


中文翻译：

---

基于钆塞酸增强 MRI 的半自动逐步专家引导 LI-RADS 分级系统


目的：肝脏影像报告和数据系统（LI-RADS）分类，特别是LR-3至5个病变与肝细胞癌（HCC）概率的识别，对于治疗策略的确定具有重要意义。我们的目标是使用深度卷积神经网络 (CNN) 开发一种针对多相钆塞酸增强 MRI 的半自动 LI-RADS 分级系统。

患者和方法：纳入了 439 名患者的内部数据集和 71 名疑似 HCC 患者的外部数据集，并接受了钆塞酸增强 MRI。专家指导的 LI-RADS 分级系统由四个深度 3D CNN 模型组成，包括用于自动直径估计的肿瘤分割模型和 LI-RADS 主要特征的三个分类模型，包括动脉期超增强 (APHE)、冲洗和增强胶囊。为了进行比较，开发了一个由单个深度 CNN 模型组成的端到端学习系统，该模型可以直接对 LI-RADS 等级进行分类。

结果：在内部测试集上，分割模型的平均骰子数为 0.84，映射直径区间的准确度为 82.7%（95% CI：74.4%、91.7%）。 APHE、洗脱剂和胶囊的曲线下面积 (AUC) 分别为 0.941 (95% CI: 0.914, 0.961)、0.859 (95% CI: 0.823, 0.890) 和 0.712 (95% CI: 0.668, 0.754)。专家指导系统的性能显着优于端到端系统，LI-RADS 分级准确度为 68.3%（95% CI：60.8%、76.5%）vs 55.6%（95% CI：48.8%、63.0%）（ P < 0.0001）。在外部测试集上，映射直径区间的准确度为 91.5%（95% CI：81.9%、100.0%）。 APHE、洗脱剂和胶囊的 AUC 分别为 0.792 (95% CI: 0.745, 0.833)、0.654 (95% CI: 0.602, 0.703) 和 0.658 (95% CI: 0.606, 0.707)。专家指导系统的总体评分准确率为66.2%（95% CI：58.0%、75.2%），显着高于端到端系统的50.1%（95% CI：43.1%、58.1%）（ P < 0.0001）。

结论：我们开发了一种半自动分步专家指导的 LI-RADS 评分系统（LR-3 至 5），优于传统的端到端学习系统。这种基于深度学习的系统可以提高临床实践中 HCC 诊断的工作流程效率。


关键词：肝脏影像报告和数据系统，LI-RADS，肝细胞癌，HCC，磁共振成像，MRI，深度学习