Predicting Treatment Response to Transcatheter Arterial Chemoembolization in Hepatocellular Carcinoma Patients using a Deep Learning-Based Approach

Zhi-Wei Li; Chun-Wang Yuan; Jian Wei; Da-Wei Yang; Hui Xu; Ying Chen; Wei Ma; Zhen-Chang Wang; Zheng-Han Yang; A-Hong Ren

doi:10.2174/0115734056367143250610045305

Predicting Treatment Response to Transcatheter Arterial Chemoembolization in Hepatocellular Carcinoma Patients using a Deep Learning-Based Approach

Curr Med Imaging. 2025 Jun 20. doi: 10.2174/0115734056367143250610045305. Online ahead of print.

Authors

Zhi-Wei Li^{1

2}, Chun-Wang Yuan³, Jian Wei⁴, Da-Wei Yang¹, Hui Xu¹, Ying Chen⁵, Wei Ma⁵, Zhen-Chang Wang¹, Zheng-Han Yang¹, A-Hong Ren¹

Affiliations

¹ Department of Radiology, Beijing Friendship Hospital, Capital Medical University, 95 YongAn Road, Xicheng District, Beijing, 100050, P.R. China.
² Department of Radiology, Beijing Jianjia Rehabilitation Hospital, Building 3, Yard 6, Kangxin Road, Fengtai District, Beijing, 100071, P.R. China.
³ Center of Interventional Oncology and Liver Diseases, Beijing Youan Hospital, Capital Medical University; No.8 Xitoutiao, Youwai St, Fengtai District, Beijing, 100069, P.R. China.
⁴ Department of Interventional Radiography, Beijing Friendship Hospital, Capital Medical University, 95 YongAn Road, Xicheng District, Beijing, 100050, P.R. China.
⁵ Faculty of Information Technology, Beijing University of Technology, Beijing, 100124, P.R. China.

PMID: 40574392
DOI: 10.2174/0115734056367143250610045305

Abstract

Objectives: This study aimed to assess the effectiveness and precision of a deep learning-based model in forecasting the early response of HCC patients to TACE.

Methods: A comprehensive review of HCC-TACE data involving 111 patients with HCC was carried out, encompassing both pre-TACE MR images (captured before the first TACE) and post-TACE imaging (acquired between 30 and 60 days following TACE). Based on the mRECIST criteria, patients were divided into two cohorts: a training dataset (91 subjects, 645 images) and a test dataset (20 subjects, 155 images). A deep learning based model utilizing LeNet architecture with an attention mechanism was developed, targeting the prediction of HCC patients' response to TACE. The robustness and accuracy of the model were examined via ROC curves and confusion matrices.

Results: Post-TACE treatment, 56 patients (50.5%) manifested an objective response (CR+PR), whereas 55 patients (49.5%) exhibited no response (SD+PD). Concerning the model's predictive ability for TACE response, the AUC was found to be 0.760 in the training dataset and 0.729 in the test dataset. The model's prediction accuracy was further corroborated by the confusion matrix, revealing an average accuracy of 70.7% in the training dataset and 72.3% in the test dataset.

Conclusion: Implementing a deep learning-based model using MRI data is potent for forecasting HCC patients' response to TACE treatment. The novel LeNet model with the attention mechanism conceived in this study contributes valuable insights that can guide the formulation of effective treatment strategies.

Keywords: Deep learning; Magnetic Resonance Imaging (MRI); Modified Response Evaluation Criteria in Solid Tumors (mRECIST); Prediction; Receiver Operating Characteristic (ROC)..