摘要: |
目的 利用Logistic回归模型评价常规超声联合声触诊组织成像定量技术(VTIQ)对三阴性乳腺癌(TNBC)的诊断价值。方法 回顾性分析170例乳腺癌患者的术前常规超声及VTIQ的声像图特征,应用曲线下面积(AUC)比较分析VTIQ各参数诊断TNBC的价值, 筛选出诊断TNBC最有价值的指标, 并建立常规超声联合VTIQ诊断TNBC的二分类Logistic回归模型。结果 TNBC和non-TNBC的最大SWV值、平均SWV值、最小SWV值、平均SWV值/正常乳腺值两组间比较差异均有统计学意义(P均<0.01), 根据AUC显示最大SWV值的诊断效能最佳。多因素Logistic回归分析后显示诊断TNBC的超声特征为边缘不光整伴分叶、后方回声增强、形态规则、最大SWV值<6.98m/s。该Logistic模型具有统计学意义(2=68.128,P<0.01),该模型的敏感度、特异度、阳性预测值、阴性预测值分别为80.6%、85.7%、72.0%、87.6%。结论 常规超声联合VTIQ的Logistic回归模型对TNBC有较高的诊断效能,可作为预测TNBC的一项重要参考指标。 |
关键词: Logistic回归模型 声触诊组织成像定量技术 三阴性乳腺癌 |
DOI: |
投稿时间:2021-10-04修订日期:2021-10-04 |
基金项目:浙江省医药卫生科技计划项目 (2020KY852) 浙江省基础公益研究项目(LQ19H180003) |
|
Wang Jue,Chen Saijun,Zhan Jinyong,Xu Kaiying,Wei Xiuzhi,Ben ZhifeiDepartment of Ultrasound, HwaMei Hosptial,University Of Chinese Academy Of Sciences ,Ningbo 315000, China |
Chen Saijun,Zhan Jinyong,Xu Kaiying,Wei Xiuzhi,Ben Zhifei |
() |
Abstract: |
ABSTRACT Objective To evaluation of conventional ultrasound and virtual touch tissue imaging
quantification(VTIQ) technology in the diagnosis of triple negative breast cancer(TNBC). Methods Conventional ultrasound and VTIQ were performed in 170 patients,ROC curves were used to compare the value of each indicator of VTIQ in diagnosing TNBC and to obtain the most valuable indicator.The ultrasonographic features of conventional ultrasound combine with VTIQ were summarized to screen out the most valuable indicators in diagnosing TNBC. Results The the differents of the SWVmax、SWVmean、SWVmin、SWVmean/mon of TNBC and non-TNBC were statistically different(all P<0.01), according to the area under the curve (AUC), the most valuable indicator is SWVmax. Edge not Smoothing with phyllodes,posteriorenhancement,regular morphology, SWVmax<6.98m/s were finally entered into the logistic regression model to diagnosis TNBC.The Logistic model has statistical significance(2=68.128,P<0.01),the sensitivity,specificity,positive predictive value and negative predictive value were 80.6%、85.7%、72.0%、87.6%,respectively. Conclusions The Logistic regression model of conventional ultrasound combined with VTIQ has a high diagnostic efficiency for TNBC,it can be used as an important reference index to predict TNBC. |
Key words: Logistic regression Virtual touch tissue imaging quantification Triple negative breast cancer |