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

作　　者：范嘉玮 程梦雨 杨哲 李文菲[1] 赵月梅 王占秋[1] FAN Jia-wei;CHENG Meng-yu;YANG Zhe;LI Wen-fei;ZHAO Yue-mei;WANG Zhan-qiu(Department of Radiology,the First Hospital of Qinhuangdao,Qinhuangdao Hebei 066000,China)

机构地区：[1]秦皇岛市第一医院影像科,河北秦皇岛066000

出　　处：《中国临床医学影像杂志》2024年第1期26-30,45,共6页Journal of China Clinic Medical Imaging

摘　　要：目的:研究增强CT影像组学术前预测喉鳞癌术后复发的价值。方法:收集本院64例经术后病理证实为喉鳞癌的患者作为训练组;将TCGA数据库中下载的39例患者作为验证组,基于患者术前CT增强静脉期图像,利用3D-slicer软件提取喉鳞癌全域影像组学特征。在训练组中,应用单因素COX及LASSO回归确定复发相关的最佳组学特征,构建影像组学风险评分对喉鳞癌患者预后分层;基于多因素COX回归构建预测1年、3年无进展生存的列线图。采用受试者工作特征曲线(ROC)评估影像组学风险评分的诊断效能。最后采用相同阈值对验证组患者进行模型验证。结果:训练组确定4个影像组学特征与复发显著相关,构建影像组学评分模型,Kaplan-Meier曲线显示训练组和验证组低风险组(组学评分<0.7)无进展生存时间均显著长于高风险组(P<0.001,P=0.025)。多因素COX回归研究表明影像组学评分是喉鳞癌术后复发的独立危险因素。在训练组中列线图预测喉鳞癌患者三年术后无进展生存C指数为0.816,影像组学模型C指数为0.795。在验证组中列线图C指数为0.858,影像组学模型C指数为0.795。结论:基于术前CT增强图像提取的喉鳞癌组学特征并建立列线图模型,可用于术前预测患者术后复发,且诊断效能高于临床常用TNM分期,可作为临床诊疗决策依据。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 optimum 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 patients 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 progression-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 treatm

关 键 词：喉肿瘤 体层摄影术 螺旋计算机 

分 类 号：R739.65[医药卫生—肿瘤] R730.261[医药卫生—临床医学] R814.42