Logistic regression analysis of ultrasonographic features in thyroid solitary nodular
- VernacularTitle:甲状腺单发结节超声特征的Logistic回归分析
- Author:
Zhanwu FENG
;
Shuzhen CONG
;
Kang LI
;
Lisang WU
;
Qing CHEN
;
Kehong GAN
- Publication Type:Journal Article
- Keywords:
Thyroid nodule;
Ultrasonography;
Logistic regression
- From:
Chinese Journal of Medical Imaging Technology
2010;26(1):66-68
- CountryChina
- Language:Chinese
-
Abstract:
Objective To find out the valuable ultrasonographic features for differentiating benign and malignant thyroid solitary nodular, and to apply the binary Logistic regression model in analysis of ultrasonography of thyroid solitary nodular. Methods Two-dimensional ultrasonography was performed in 194 patients with thyroid solitary nodular confirmed with surgical pathology. A Logistic model was obtained on the basis of ultrasonographic features. A receiver operator characteristic (ROC) curve was constructed to assess the performance of the Logistic model. Results Three ultrasonographic features including shape, calcification and heterogeneous texture were finally entered into the Logistic model. The percentage of correct prediction was 91.75%. The area under ROC curve was 0.916±0.035. Conclusion The binary Logistic regression can select the valuable indexes in the differential diagnosis of thyroid solitary nodular. The application of binary Logistic regression model can improve the diagnosis accuracy of thyroid solitary nodular.