首页 / 院系成果 / 成果详情页

Integrating computed tomography image features improves clinical prediction models for outcomes in nasopharyngeal carcinoma patients treated with (chemo)radiation 期刊论文

编号：

0A72BFD64F2B012C2BC776AC2FC71C02
作者：

Zhou, Guanzhi^[1,2] Ma, Baoqiang^[1,3] Li, Yan^[1,4] Yang, Pei^[2] Shi, Yingrui^[5,6] van der Schaaf, Arjen^[1] van Dijk, Lisanne V.^[1] Langendijk, Johannes A.^[1] Sijtsema, Nanna M.^[1]
语种：

英文
期刊：

PHYSICS & IMAGING IN RADIATION ONCOLOGY ISSN：2405-6316 2026 年 40 卷 ; JUL
收录：
关键词：

Nasopharyngeal carcinoma computed tomography prognosis treatment outcome radiomics deep-learning
摘要：

Background and purpose: Clinical prognostic models for nasopharyngeal carcinoma (NPC) treated with intensitymodulated radiotherapy (IMRT) with or without chemotherapy remain insufficient to capture tumour heterogeneity. We investigated whether computed tomography (CT)-based signatures add prognostic value for overall survival, progression-free survival, local control and distant control in NPC patients. Materials and methods: The study population consisted of 1360 patients with stage I-IVa NPC treated with (chemo)IMRT (2013-2017). Radiomic and deep-learning features were analysed with twelve clinical variables. Radiomic models were built using bootstrap resampling feature selection and multivariable Cox regression; deeplearning models used 3D ResNet-18 or DenseNet-121. Models were evaluated on an internal hold-out test set (n = 409; training set n = 951) with the concordance index and compared against clinical-only reference models. Decision curve analysis was used to assess clinical utility. Results: Adding radiomic primary tumour features (Neighbouring Gray Tone Difference Matrix - coarseness) improved local control concordance index from 0.51 to 0.60 (p = 0.02). A DenseNet-121 combining clinical data with composite primary tumour and lymph node masks achieved the highest distant control (0.68 vs 0.66, p = 0.01). For overall survival and progression-free survival, the improvements were not significant. Decision curve analysis demonstrated net benefit of the DenseNet-121 distant control model over treat-all and treat-none strategies at threshold probabilities of 10-25%. Conclusions: Incorporating CT-based radiomic and deep-learning features into prognostic models significantly improved prediction of local and distant control in NPC, supporting their potential as imaging biomarkers for refined risk stratification.
推荐引用方式
GB/T 7714：

Zhou Guanzhi,Ma Baoqiang,Li Yan, et al. Integrating computed tomography image features improves clinical prediction models for outcomes in nasopharyngeal carcinoma patients treated with (chemo)radiation [J].PHYSICS & IMAGING IN RADIATION ONCOLOGY,2026,40.
APA：

Zhou Guanzhi,Ma Baoqiang,Li Yan,Yang Pei,&Sijtsema Nanna M..(2026).Integrating computed tomography image features improves clinical prediction models for outcomes in nasopharyngeal carcinoma patients treated with (chemo)radiation .PHYSICS & IMAGING IN RADIATION ONCOLOGY,40.
MLA：

Zhou Guanzhi, et al. "Integrating computed tomography image features improves clinical prediction models for outcomes in nasopharyngeal carcinoma patients treated with (chemo)radiation" .PHYSICS & IMAGING IN RADIATION ONCOLOGY 40(2026).
入库时间：

2026/8/12 21:51:06
更新时间：

2026/8/12 21:51:06

浏览次数：1  下载次数：0

分享到：

浏览次数：1

下载次数：0

打印次数：0