医疗卫生装备2025，Vol.46Issue(11)：50-56,7.DOI:10.19745/j.1003-8868.2025200
瘤内和瘤周影像组学特征在预测肝细胞癌放疗疗效中的应用价值
Value of intratumor and peritumor imaging radiomics features in predicting efficacy of radiotherapy for hepatocellular carcinoma
摘要
Abstract
Objective To develop a fused model based on combined intratumor-peritumor radiomics features and clinical risk factors to accurately predict the efficacy of radiotherapy for hepatocellular carcinoma(HCC).Methods A retrospective study was conducted on 124 HCC patients(95 males and 29 females)undergoing IMRT at the radiotherapy center of some hospital,who were divided into a training set and a validation set in a 7∶3 ratio.According to relevant standards for evaluating solid tumor treatment efficacy the patients at the progressive disease phase were enrolled into a non-response group,and the ones at the phases of complete response,partial response and stable disease were included into a response group.Firstly,radiomics features were extracted from pre-radiotherapy contrast-enhanced CT images,and were screened with the least absolute shrinkage and selection operator(LASSO)regression to obtain valuable radiomics features;secondly,the scores of intratumor radiomics features,peritumor radiomics features and combined intratumor-peritumor radiomics features underwent binary Logistic regression(LR)analysis by using the glmnet package in R software(version 4.2.0),so as to construct three models for intratumor radiomics features,peritumor radiomics features and combined intratumor-peritumor radiomics features;finally,a fused model was established based on the scores of combined intratumor-peritumor radiomics features and clinical risk factors,and the corresponding Nomogram,calibration curve and clinical decision curve were plotted.R software(version 4.4.1)was used for image analysis and statistical analysis.Results The fused model based on the scores of combined intratumor-peritumor radiomics features and clinical risk factors achieved an AUC of 0.87 on both the training and validation sets,outperforming the models based solely on intratumor radiomics features,peritumor radiomics features or combined intratumor-peritumor radiomics features.The Nomogram plot showed the prediction probability of the fused model for the short-term treatment response rate was 10%-90%within the total score range of 42-93 points,and the calibration and clinical decision curves indicated that the fused model gained advantages in prediction and clinical application.Conclusion The fused model based on the scores of combined intratumor-peritumor radiomics features and clinical risk factors predicts the short-term efficacy of HCC radiotherapy effectively,providing references for formulating radiotherapy plans and guiding subsequent treatment decisions.关键词
影像组学/瘤周/瘤内/HCC/原发性肝癌/IMRT/疗效预测Key words
radiomics/peritumor/intratumor/hepatocellular carcinoma/primary hepatic carcinoma/intensity-modulated radiation therapy/therapeutic efficacy prediction分类
基础医学引用本文复制引用
杜星宇,李金凯,唐媛媛,许晓燕,刘希胜..瘤内和瘤周影像组学特征在预测肝细胞癌放疗疗效中的应用价值[J].医疗卫生装备,2025,46(11):50-56,7.基金项目
国家自然科学基金面上项目(82073344) （82073344）
江苏省放射治疗学医学重点学科项目(JSDW202238) （JSDW202238）
江苏省人民医院临床能力提升工程项目(JSPH-MA-2023-6) （JSPH-MA-2023-6）
