| 摘要: |
| 目的 探讨乳腺小肿物的超声征象,评价其应用价值。 方法 回顾性分析了通过手术病理证实的最大径≤10.0mm乳腺肿物(其中良性466个,恶性31个)的超声征象,通过多因素二元Logistic回归分析筛选出乳腺小肿物的恶性超声征象,并建立了回归模型。 结果 多因素二元Logistic回归分析显示最后进入回归模型的是乳腺小肿物的方向和血流信号。Logistic回归模型以P=0.05为阈值,准确率达95.2%。结论 以方向和血流信号2个诊断参数变量建立的Logistic回归模型有助于乳腺小肿物的鉴别诊断。 |
| 关键词: 彩色多普勒超声 乳腺小肿物 多因素二元Logistic回归 |
| DOI： |
| 投稿时间：2019-03-10修订日期：2019-03-21 |
| 基金项目: |
|
| The multi-factor binary Logistic regression analysis of subcentimeter breast masswith color-doppler ultrasound |
| Li Yu-sheng,Wei Xiu-xia |
| () |
| Abstract: |
| Object To approach the sonographic features of subcentimeter breast mass and assess the diagnostic value of sonographic findings of subcentimeter breast mass. Methods The sonographic characteristics of 491 breast mass(≤10.0mm) were reviewed retrospectively．The multi-factor binary Logistic regression analysis for predicting the subcentimeter breast mass. Results The shape and vascularity were finally entering the Logistic model. If apply =0.05 as cut-off, the diagnostic accuracy was 95.2%. Conclusion The Logistic model which is based on shape and vascularity can efficiently differentiate malignant subcentimeter breast mass from benign ones. |
| Key words: color-doppler ultrasound, subcentimeter breast mass, multi-factor binary Logistic regression analysis |
