一种结合多模态因素预测经动脉化疗栓塞术总生存期的深度学习模型

A Deep Learning Model Combining Multimodal Factors to Predict the Overall Survival of Transarterial Chemoembolization.

作者信息

Sun Zhongqi, Li Xin, Liang Hongwei, Shi Zhongxing, Ren Hongjia

机构信息

Department of Radiology, the Second Affiliated Hospital of Harbin Medical University, Harbin, 150086, People's Republic of China.

Department of Interventional Radiology, the Second Affiliated Hospital of Harbin Medical University, Harbin, 150086, People's Republic of China.

出版信息

J Hepatocell Carcinoma. 2024 Feb 26;11:385-397. doi: 10.2147/JHC.S443660. eCollection 2024.

BACKGROUND

To develop and validate an overall survival (OS) prediction model for transarterial chemoembolization (TACE).

METHODS

In this retrospective study, 301 patients with hepatocellular carcinoma (HCC) who received TACE from 2012 to 2015 were collected. The residual network was used to extract prognostic information from CT images, which was then combined with the clinical factors adjusted by COX regression to predict survival using a modified deep learning model (DLOP). The DLOP model was compared with the residual network model (DLOP), multiple COX regression model (DLOP), Radiomic model (Radiomic), and clinical model.

RESULTS

In the validation cohort, DLOP shows the highest TD AUC of all cohorts, which compared with Radiomic (TD AUC: 0.96vs 0.63) and clinical model (TD AUC: 0.96 vs 0.62) model. DLOP showed significant difference in C index compared with DLOP and DLOP models ( < 0.05). Moreover, the DLOP showed good calibration and overall net benefit. Patients with DLOP model score ≤ 0.902 had better OS (33 months vs 15.5 months, < 0.0001).

CONCLUSION

The deep learning model can effectively predict the patients' overall survival of TACE.

背景

开发并验证经动脉化疗栓塞术（TACE）的总生存期（OS）预测模型。

方法

在这项回顾性研究中，收集了2012年至2015年接受TACE治疗的301例肝细胞癌（HCC）患者。使用残差网络从CT图像中提取预后信息，然后将其与通过COX回归调整后的临床因素相结合，使用改良的深度学习模型（DLOP）预测生存期。将DLOP模型与残差网络模型（DLOP）、多重COX回归模型（DLOP）、影像组学模型（Radiomic）和临床模型进行比较。

结果

在验证队列中，DLOP在所有队列中显示出最高的TD AUC，与影像组学模型（TD AUC：0.96对0.63）和临床模型（TD AUC：0.96对0.62）相比。与DLOP和DLOP模型相比，DLOP在C指数上显示出显著差异（<0.05）。此外，DLOP显示出良好的校准和总体净效益。DLOP模型评分≤0.902的患者OS更好（33个月对15.5个月，<0.0001）。