摘要
目的探讨MRI影像组学参数预测宫颈鳞癌p53的价值。资料与方法回顾性分析53例宫颈鳞癌患者的MRI图像,通过组学软件提取影像组学特征,选取术前MRI影像组学所勾画的肿瘤区域行免疫组化染色,根据术后病理免疫组化p53表达结果分为阳性32例和阴性21例。分析两组患者细胞分化程度的差异和两组患者396个影像组学数据特征的差异,采用受试者工作特征(ROC)曲线确定各特征参数预测宫颈鳞癌p53的敏感度及特异度。结果p53阳性组中低分化鳞癌占75.0%(24/32),明显高于p53阴性组的52.3%(11/21),差异有统计学意义(P=0.042)。两组患者MRI影像组学的4个特征最大强度、灰度值范围、反转分矩、短运行强度差异均有统计学意义(P<0.05),这4个影像组学特征的ROC曲线敏感度分别为81%、75%、50%和75%,特异度分别为52%、57%、90%和67%。结论MRI影像组学特征最大强度、灰度值范围、反转分矩、短运行强度对预测宫颈鳞癌组织p53表达有较大的临床价值。
Purpose To explore the value of MRI radiomics parameters in predicting p53 of cervical squamous cell carcinoma.Materials and Methods MR images of 53 patients with cervical squamous cell carcinoma were retrospectively analyzed.The radiomics parameters were extracted with radiomics software and tumor region outlined with preoperative MRI radiomics was chosen for immunohistochemical staining.All patients were divided into p53 positive(32 cases)and p53 negative(21 cases)group according to postoperative pathologic immunohistologic p53 expression.The difference in cell differentiation and differences of characteristics of 396 radiomics data of patients between the two groups were analyzed.The receiver operating characteristic(ROC)curve was used to determine the sensitivity and specificity of each parameter for predicting p53 in cervical squamous cell carcinoma.Results The poorly differentiated squamous cell carcinoma in p53-positive group accounted for 75.0%(24/32),which was significantly higher than that in patients with poorly differentiated squamous cell carcinoma in p53-negative group(52.3%,11/21,P=0.042).Four MRI imaging radiomics characteristics(maximum intensity,gray value range,reversal moment and short running intensity)of patients between the two groups were of statistical significance(P<0.05).Sensitivity of ROC curve for the four imaging radiomics characteristics was 81%,75%,50%and 75%,respectively;the specificity was 52%,57%,90%and 67%,respectively.Conclusion MRI imaging radiomics characteristics(maximum intensity,gray value range,reversal moment and short running intensity)are of greater clinical value in predicting expression of p53 of cervical squamous cell carcinoma.
作者
韦明珠
赵振华
胡红杰
章俞
毛海佳
黄亚男
WEI Mingzhu;ZHAO Zhenhua;HU Hongjie;ZHANG Yu;MAO Haijia;HUANG Yanan(Department of Radiology,Sir Run Run Shaw Hospital,School of Medicine,Zhejiang University,Hangzhou 310016,China;不详)
出处
《中国医学影像学杂志》
CSCD
北大核心
2019年第12期934-937,947,共5页
Chinese Journal of Medical Imaging
基金
浙江省医药卫生科技计划项目（2015KYB402）
浙江省医药卫生科研基金项目（2017KY663）
关键词
宫颈肿瘤
癌
鳞状细胞
磁共振成像
基因
p53
Uterine cervical neoplasms
Carcinoma,squamous cell
Magnetic resonance imaging
Genes,p53
