中国临床医学影像杂志2024，Vol.35Issue(1)：26-30,45,6.DOI:10.12117/jccmi.2024.01.006
术前静脉期CT影像组学列线图预测喉鳞癌患者预后
Preoperative venous CT radiomics model in predicting the prognosis of laryngeal squamous cell carcinoma patients
范嘉玮 1程梦雨 1杨哲 1李文菲 1赵月梅 1王占秋1
作者信息
- 1. 秦皇岛市第一医院影像科,河北 秦皇岛 066000
- 折叠
摘要
Abstract
Objective:To explore the value of preoperative contrasted CT radiomics in predicting postoperative recurrence of laryngeal squamous cell carcinoma(LSCC).Methods:Sixty-four patients with LSCC confirmed by postoperative pathology in our hospital were collected as the training group;39 patients downloaded from the TCGA database were used as the validation group.3D-slicer software was performed to extract the global radiomics features of laryngeal cancer based on the patients'preoperative CT-enhanced venous phase images.Univariate COX and LASSO regression were applied to determine the opti-mum radiomics features related to recurrence in the training set,radiomics risk score was constructed to stratify the patients with LSCC.Nomograms were constructed based on multivariate COX analysis to predict progression-free survival at 1 and 3 years.The diagnostic performance of the radiomics risk score was assessed by receiver operating characteristic(ROC).Finally,the same threshold was used to validate the model for the validation group of patients.Results:Four radiomics features were significantly associated with recurrence,and developed a radiomics risk score model.The Kaplan-Meier curve showed that pa-tients in low-risk group(Risk score=0.7)favor with longer progression-free survival time than those in the high-risk group in both the training and the validation groups(P<0.001,P=0.025).Multivariate COX regression study showed that radiomics risk score was an independent risk factor for postoperative recurrence of LSCC.The C index of nomogram for predicting the pro-gression-free survival of three years in the training set was 0.816,and the C index of radiomics model was 0.795.And the C index of nomogram was 0.858,the C index of radiomics model was 0.795 in the validation set,respectively.Conclusion:The radiomics features derived from preoperative CT enhanced images can be used to predict postoperative recurrence with better diagnostic efficiency than clinical TNM staging,which can be used to guide clinical diagnosis and treatment decisions.关键词
喉肿瘤/体层摄影术,螺旋计算机Key words
Laryngeal Neoplasms/Tomography,Spiral Computed分类
医药卫生引用本文复制引用
范嘉玮,程梦雨,杨哲,李文菲,赵月梅,王占秋..术前静脉期CT影像组学列线图预测喉鳞癌患者预后[J].中国临床医学影像杂志,2024,35(1):26-30,45,6.
