文章摘要
米晓然,王金萍,刘莎,等.基于甲状腺细针穿刺针感及临床超声特征的列线图模型预测甲状腺 BSRTC Ⅲ类结节的良恶性[J].安徽医药,2026,30(2):352-358.
基于甲状腺细针穿刺针感及临床超声特征的列线图模型预测甲状腺 BSRTC Ⅲ类结节的良恶性
A nomogram model based on thyroid fine needle puncture needle sense and clinical ultrasound features predicted the benign and malignant thyroid BSRTC class Ⅲ nodules
  
DOI:10.3969/j.issn.1009-6469.2026.02.027
中文关键词: 甲状腺结节  活组织检查,针吸  穿刺针感  列线图
英文关键词: Thyroid nodule  Biopsy, needle  Puncture feeling  Nomogram
基金项目:中国超声医师科技新星计划课题( KJXX2021001);山西省重点医学科研项目( 2023XM038)
作者单位E-mail
米晓然 山西医科大学医学影像学院,山西 太原 030001  
王金萍 山西医科大学第五临床医学院超声医学科,山西太原 030001  
刘莎 山西医科大学第五临床医学院超声医学科,山西太原 030001  
尉志红 山西医科大学第五临床医学院超声医学科,山西太原 030001  
严继萍 山西医科大学第五临床医学院超声医学科,山西太原 030001 13593157054@163.com 
摘要点击次数: 458
全文下载次数: 178
中文摘要:
      目的基于甲状腺细针穿刺针感及临床超声特征的列线图模型预测甲状腺 Bethesda甲状腺细胞病理学报告系统(BSRTC)Ⅲ类结节的良恶性。方法回顾分析 2018年 7月至 2023年 4月山西医科大学第五临床医学院行超声引导下细针穿刺细胞学检查,提示 BSRTC Ⅲ类且经手术病理证实的 396例甲状腺结节病人,共 396个结节,其中良性 131例,恶性 265例。将病人采用随机数字表法分为训练集( n=278)和测试集( n=118)两组,比例为 7∶3。使用最小绝对收缩与选择算子( LASSO)方法与多因素 logistic回归筛选 BSRTC Ⅲ类甲状腺恶性结节的独立预测因素,基于独立预测因素构建模型,选择最佳诊断效能的预测模型绘制列线图,最后评价该模型的一致性与临床效益。结果多因素 logistic回归分析提示年龄、结节长径、实性,纵横比≥1、有其他可疑恶性病灶、质地硬、沙砾感是甲状腺 BSRTC Ⅲ类恶性结节的独立预测因素( OR=0.95、3.96、2.58、3.77、3.27、 5.20,P<0.05)。基于独立预测因素,构建穿刺针感模型、临床超声特征模型及联合预测模型。在训练集中,联合模型( AUC=0.86)的诊断效能优于穿刺针感模型( AUC=0.68)及临床超声特征模型( AUC=0.81)。校准曲线显示预测值与实测值的一致性较高,临床决策曲线分析显示,列线图模型在所有模型中临床适用性最佳。结论甲状腺细针穿刺针感有助于甲状腺 BSRTC Ⅲ类结节良恶性的诊断,联合模型在 BSRTC Ⅲ类甲状腺结节诊断中具有较高的诊断效能,有望为临床治疗及决策提供依据。
英文摘要:
      Objective To predict the diagnostic value of benign and malignant thyroid BSRTC Ⅲ nodules based on the nomogrammodel of thyroid fine needle puncture sensation and clinical ultrasonic features. Methods A retrospective analysis of 396 patientswith BSRTC Ⅲ thyroid nodules confirmed by surgery and pathology in the Fifth Clinical College of Shanxi Medical University from Ju-ly 2018 to April 2023 was conducted. There were a total of 396 nodules, of which 131 were benign and 265 were malignant. There wasa 7:3 ratio between the training set (n=278) and the test set (n=118). Based on the training set, the least absolute shrinkage and selec-tion operator (LASSO) method were used for factor selection, and multivariate logistic regression analysis were used to screen indepen-dent risk factors to construct a prediction model, predict the malignant risk of thyroid BSRTC Ⅲ nodule, and evaluate the applicationvalue of the nomogram through internal verification.Results Multivariate logistic regression analysis indicated that age, maximum di-ameter of nodule, solid, aspect ratio ≥1, other suspected malignant lesions, hard texture and grittiness were independent risk factors forBSRTC Ⅲ malignant nodule of thyroid gland (OR=0.95,3.96,2.58,3.77,3.27,5.20) (all P<0.05).Based on independent risk factors, punc-ture needle sensation model, clinical ultrasound feature model and combined prediction model were constructed. In the training set, thediagnostic efficiency of the joint model (AUC=0.86) was better than that of the puncture needle sensation model (AUC=0.68) and theclinical ultrasonic feature model (AUC=0.81). Analysis of clinical decision curve showed that the nomogram model had the best clinicalapplicability among all models.Conclusions Thyroid puncture needle sensation is helpful for the diagnosis of benign and malignantBSRTC Ⅲ nodules. The joint model has high diagnostic efficacy in the diagnosis of BSRTC Ⅲ thyroid nodules, which is expected toprovide evidence for clinical treatment and decision-making.
查看全文   查看/发表评论  下载PDF阅读器
关闭

分享按钮