Zhou Jing, Yang Daofeng, Tang Hao
Department of Infectious Diseases, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China.
Department of Radiology, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China.
Heliyon. 2025 Jan 7;11(1):e41735. doi: 10.1016/j.heliyon.2025.e41735. eCollection 2025 Jan 15.
Previous studies mostly use single-type features to establish a prediction model. We aim to develop a comprehensive prediction model that effectively identify patients with poor prognosis for single hepatocellular carcinoma (HCC) based on artificial intelligence (AI). : 236 single HCC patients were studied to establish a comprehensive prediction model. We collected the basic information of patients and used AI to extract the features of magnetic resonance (MR) images.
The clinical model based on linear regression (LR) algorithm (AUC: 0.658, 95%CI: 0.5021-0.8137), the radiomics model and deep transfer learning (DTL) model based on light gradient-boosting machine (Light GBM) algorithm (AUC: 0.761, 95%CI: 0.6326-0.8886 and AUC: 0.784, 95%CI: 0.6587-0.9087, respectively) were the optimal prediction models. A comparison revealed that the integrated nomogram had the largest area under the receiver operating characteristic curve (AUC) (all P < 0.05). In the training cohort, the integrated nomogram was predictive of recurrence-free survival (RFS) as well as overall survival (OS) (C-index: 0.735 and 0.712, P < 0.001). In the test cohort, the integrated nomogram also can predict RFS and OS (C-index: 0.718 and 0.740, P < 0.001) in patients.
The integrated nomogram composed of signatures in the prediction models can not only predict the postoperative recurrence of single HCC patients but also stratify the risk of OS after the operation.
以往研究大多使用单一类型特征来建立预测模型。我们旨在基于人工智能（AI）开发一种综合预测模型，以有效识别单发性肝细胞癌（HCC）预后不良的患者。对236名单发性HCC患者进行研究以建立综合预测模型。我们收集了患者的基本信息，并使用AI提取磁共振（MR）图像的特征。
基于线性回归（LR）算法的临床模型（AUC：0.658，95%CI：0.5021 - 0.8137）、基于轻梯度提升机（Light GBM）算法的影像组学模型和深度迁移学习（DTL）模型（AUC分别为：0.761，95%CI：0.6326 - 0.8886和0.784，95%CI：0.6587 - 0.9087）是最优预测模型。比较显示，综合列线图在受试者操作特征曲线下面积（AUC）最大（所有P < 0.05）。在训练队列中，综合列线图可预测无复发生存期（RFS）以及总生存期（OS）（C指数：0.735和0.712，P < 0.001）。在测试队列中，综合列线图也能预测患者的RFS和OS（C指数：0.718和0.740，P < 0.001）。
由预测模型中的特征组成的综合列线图不仅可以预测单发性HCC患者的术后复发，还可以对术后OS风险进行分层。
