CT直方图分析预测十二指肠乳头癌淋巴结转移的临床应用价值  

Clinical application of CT histogram analysis in predicting lymph node metastasis in duodenal papillary carcinoma

作　　者：谢群 毕蕾 张龄艺 徐鸿志 蔡世峰 XIE Qun;BI Lei;ZHANG Lingyi;XU Hongzhi;CAI Shifeng(Department of Medical Imaging,Shandong Provincial Hospital Affiliated to Shandong First Medical University,Jinan 250021,China;Department of Medical Imaging,Linyi People’s Hospital,Linyi 276000,China;Department of Medical Ultrasound,Linyi People’s Hospital,Linyi 276000,China)

机构地区：[1]山东第一医科大学附属省立医院医学影像科,山东济南250021 [2]山东省临沂市人民医院放射科,山东临沂276000 [3]山东省临沂市人民医院超声医学科,山东临沂276000

出　　处：《医学影像学杂志》2024年第11期96-99,共4页Journal of Medical Imaging

基　　金：山东省医药卫生科技发展计划项目(编号:202209011102)

摘　　要：目的探讨CT直方图分析预测十二指肠乳头癌(DPC)淋巴结转移的临床应用价值。方法选取30例DPC患者资料,所有患者术前均行CT增强扫描。采用深睿多模态科研平台对静脉期图像勾画感兴趣区(ROI)并提取直方图参数,进行单因素分析以评估淋巴结转移阳性组和阴性组之间的参数差异。利用Pearson或Spearman的相关性分析探讨变量与淋巴结转移之间的相关性。利用受试者工作特征(ROC)曲线分析评价直方图参数对于DPC淋巴结转移的诊断效能。结果直方图参数中的均匀度、熵和方差在淋巴结转移阳性与阴性的比较均差异有统计学意义(P<0.05)。在CT直方图参数中,均匀度在预测DPC淋巴结转移时达到最高ROC曲线下面积(AUC=0.886);当均匀度和肿瘤大小相结合或均匀度与癌胚抗原(CEA)相结合时,AUC=0.915;当均匀度、肿瘤大小及CEA相结合时为最高的AUC=0.938。结论基于全病灶CT直方图分析可用于预测DPC的淋巴结转移,并且均匀度、肿瘤大小及CEA组合是区分DPC淋巴结转移的最佳选择。Objective To explore the clinical application value of CT histogram analysis in predicting lymph node metastasis in duodenal papilla carcinoma(DPC).Methods A retrospective analysis was conducted on the data of 30 patients with DPC,all of whom underwent preoperative contrast-enhanced CT scans.The Dr.Wise Multimodal Research Platform was used to delineate regions of interest on the venous phase images and extract histogram parameters.Univariate analysis was performed to evaluate the differences in parameters between the lymph node metastasis positive group and the negative group.Pearson or Spearman correlation analysis was used to explore the correlation between variables and lymph node metastasis.Logistic regression was used to find the optimal combination model.The receiver operating characteristic(ROC)curve was used to evaluate the diagnostic performance of histogram parameters for DPC lymph node metastasis.Results Histogram parameters including uniformity,entropy,and variance showed statistically significant differences between the lymph node metastasis positive group and the negative group(P<0.05).Among the CT histogram parameters,uniformity achieved the highest area under the ROC curve(AUC)in predicting lymph node metastasis of DPC,with a value of 0.886;when combined with tumor size or carcinoembryonic antigen(CEA),the AUC reached 0.915;the highest AUC of 0.938 was achieved when uniformity,tumor size,and CEA were combined.Conclusion CT histogram analysis based on the entire lesion can be used to predict lymph node metastasis of DPC,and the combination of uniformity,tumor size,and CEA is the optimal choice for distinguishing lymph node metastasis of DPC.

关 键 词：十二指肠乳头癌 直方图分析 淋巴结转移 体层摄影术 X线计算机 

分 类 号：R735.3[医药卫生—肿瘤] R814.42[医药卫生—临床医学]