基于多模态超声参数预测甲状腺微小乳头状癌颈部中央区淋巴结转移的LASSO回归分析  

LASSO Regression Analysis for Predicting Lymph Node Metastasis in the Cervical Central Region of Thyroid Micropapillary Carcinoma Based on Multimodal Ultrasound Parameters

作　　者：张清 张玉君 陶冶 胡冬梅[2] 张超学[3] ZHANG Qing;ZHANG Yujun;TAO Ye;HU Dongmei;ZHANG Chaoxue(Ultrasound Medicine Department,Taihe County People’s Hospital(Taihe Hospital Affiliated to Wannan Medical College),Anhui 236600,China;Department of Ultrasound,The Second People’s Hospital of Fuyang,Anhui 236000,China;The First Affiliated Hospital of Anhui Medical University,Anhui 230000,China)

机构地区：[1]太和县人民医院(皖南医学院附属太和医院)超声医学科,安徽236600 [2]阜阳市第二人民医院超声科,安徽236000 [3]安徽医科大学第一附属医院,安徽230000

出　　处：《影像科学与光化学》2025年第2期11-17,29,共8页Imaging Science and Photochemistry

基　　金：安徽省转化医学研究院科研基金项目(2021zhyx⁃C35)。

摘　　要：目的:本研究旨在探讨基于常规超声、剪切波弹性成像及超声造影等多模态超声参数,利用LASSO回归分析和Logistic回归构建预测甲状腺微小乳头状癌(PTMC)颈部中央区淋巴结转移(CLNM)的模型。方法:选取2022年6月至2024年6月期间收治的220例PTMC患者,根据术后病理结果分为CLNM组和非CLNM组,通过LASSO回归筛选出年龄、结节最大径、微钙化、被膜侵犯、增强均匀性和弹性评分6个关键因素,并构建预测模型。结果:该模型在预测CLNM时,AUC值达到0.825(95%CI:0.769~0.873),敏感度为81.10%,特异度为71.90%。结论:基于多模态超声特征构建的预测模型能有效预测PTMC患者的CLNM风险,为临床精准治疗方案的制定提供有力支持。Objective:This study aimed to develop a prediction model for central cervical lymph node metastasis(CLNM)in papillary thyroid microcarcinoma(PTMC)patients based on multimodal ultrasound parameters including conventional ultrasound,shear wave elastography,and contrast⁃enhanced ultrasound,using LASSO regression analysis and Logistic regression.Methods:A total of 220 PTMC patients from June 2022 to June 2024 were enrolled and divided into CLNM and non⁃CLNM groups according to postoperative pathological.Six key factors including age,nodule maximum diameter,microcalcification,capsular invasion,enhancement uniformity,and elasticity score were screened out by LASSO regression and used to construct the prediction model.Results:The model achieved an AUC value of 0.825(95%CI:0.769⁃0.873)with a sensitivity of 81.10%and a specificity of 71.90%in predicting CLNM.Conclusion:The prediction model based on multimodal ultrasound features can effectively predict the risk of CLNM in PTMC patients,providing valuable support for the formulation of precision treatment protocols in clinical practice.

关 键 词：甲状腺微小乳头状癌 颈部中央区淋巴结转移 超声 多模态 

分 类 号：R736.1[医药卫生—肿瘤]