Machine Learning Radiomics for Predicting Response to MR-Guided Radiotherapy in Unresectable Hepatocellular Carcinoma - A Multicenter Cohort Study

Aug 2, 2026

Upload by Miles

This multicenter retrospective study evaluates magnetic resonance (MR)-guided hypofractionated radiotherapy for patients with unresectable hepatocellular carcinoma (HCC) and examines a machine learning radiomics workflow for treatment response prediction. Radiomics features were derived from gross tumor volume (GTV) and reduced using K-means clustering to form two radiomics-defined subtypes. Nine machine learning models were trained and validated via 5-fold cross-validation, with internal and external validation for a multilayer perceptron model. Results show no Grade 3/4 adverse events and significantly different survival outcomes between subtypes.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages15

Reading Time38 min

AccessFree to download

Background

Methods

Results

Conclusion

Journal of Hepatocellular Carcinoma downloaded from [https://www.dovepress.com/](https://www.dovepress.com/) For personal use only. Journal of Hepatocellular Carcinoma Open Access Full Text Article ORIGINAL RESEARCH Machine Learning Radiomics for Predicting Response to MR-Guided Radiotherapy in Unresectable Hepatocellular Carcinoma: A Multicenter Cohort Study Ke Su 1–3 , *, Xin Liu 1 ,4 , *, Yue-Can Zeng 5 , *, Junnv Xu 6 , *, Han Li 2 , Heran Wang 7 , Shanshan Du 1 , Huadong Wang 1 , Jinbo Yue8 , Yong Yin 1 , Zhenjiang Li 1 1Department of Radiation Physics, Shandong Cancer Hospital and Institute, Shandong First Medical University and Shandong Academy of Medical Sciences, Jinan, 250117, People’s Republic of China; 2Department of Oncology, The Affiliated Hospital of Southwest Medical University, Luzhou, 646000, People’s Republic of China; 3Department of Radiation Oncology, National Cancer Center/ National Clinical Research Center for Cancer/ Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, 100000, People’s Republic of China; 4Department of Gynecological Radiotherapy, Harbin Medical University Cancer Hospital, Harbin, 150081, People’s Republic of China; 5Department of Radiation Oncology, Cancer Treatment Center, The Second Affiliated Hospital of Hainan Medical University, Haikou, 570311, People’s Republic of China; 6Department of Medical Oncology, The Second Affiliated Hospital, Hainan Medical University, Haikou, Hainan Province, 570311, People’s Republic of China; 7Department of Orthopedics, Shengjing Hospital of China Medical University, Shenyang, Liaoning, 110004, People’s Republic of China; 8Department of Abdominal Radiotherapy, Shandong Cancer Hospital and Institute, Shandong First Medical University and Shandong Academy of Medical Sciences, Jinan, 250117, People’s Republic of China *These authors contributed equally to this work Correspondence: Yong Yin; Zhenjiang Li, [Email yinyongsd@126.com](Email yinyongsd@126.com); zhenjli [1987@163.com](1987@163.com) | Background: This study was conducted to assess the efficacy and safety of magnetic resonance (MR)-guided hypofractionated radiotherapy in patients with unresectable hepatocellular carcinoma (HCC) . Machine learning-based radiomics was utilized to predict responses in these patients.
Methods: This retrospective study included 118 hCC patients who received MR-guided hypofractionated radiotherapy. The primary study endpoint was the objective response rate (ORR) . Radiomics features were based on the gross tumor volume (GTV) . K-means clustering was performed to differentiate cancer subtypes based on radiomics. Nine radiomics-utilizing machine learning models were built and validated internally through 5-fold cross-validation.
Results: The ORR, median progression-free survival (mPFS), and median overall survival (mOS) were 54.4%, 21.7 months, and 40.7 months, respectively. No patient experienced Grade 3/4 adverse events. 1130 radiomics features were extracted from the GTV, of which 7 were included for further analysis. K-means clustering identified 2 subtypes based on the selected features. Subtype 1 had significantly higher response, longer mPFS, and longer mOS than Subtype 2. In both internal and external validations, the multi-layer perceptron (MLP) model demonstrated superior predictive performance for response, achieving a receiver operating characteristic-area under the curve (ROC-AUC) of 0.804 and 0.842, respectively.
Conclusion: MR-guided radiotherapy was proven to be effective and safe for HCC. The machine learning radiomics model developed in this study could accurately predict the response of radiotherapy-treated inoperable HCC.
Keywords: machine learning models, radiomics, radiotherapy, hepatocellular carcinoma | | --- | | Introduction
Hepatocellular carcinoma (HCC) is the most prevalent form of primary liver cancer and poses a major global health challenge.1 It accounts for app

FAQs

What patient population and treatment were evaluated in this study?

The study included 118 patients with unresectable hepatocellular carcinoma treated with MR-guided hypofractionated radiotherapy.

How were radiomics features used to predict radiotherapy response?

Radiomics features were extracted from the GTV, clustered using K-means to differentiate tumor subtypes, and used to build and validate nine machine learning models. The multilayer perceptron showed the best predictive performance.

What were the main outcome and safety results?

The objective response rate was 54.4%, with median progression-free survival of 21.7 months and median overall survival of 40.7 months. No patients experienced Grade 3/4 adverse events.

Aug 2, 2026

Upload by Miles