Evaluation Of Ultrasonographic In Diagnosis Of Benign And Malignant Thyroid Nodules By Binary Logistic

Objective:Based on ultrasonic diagnosis of thyroid nodule as a variable of th e Logistic regression model,evaluate the value of gray scale ultrasonography(US),col or Doppler and ultrasound elastography(UE) in thyroid nodule. Methods:After sur gical resection and confirmed by pathology were retrospectively analyzed in 113 ca ses(136 lesions)and the clinical data of patients without diffuse lesions of the thyro id gland,preoperative both for gray scale ultrasonography(US),color Doppler and ultr asound elastography(UE)imaging.Conventional two-dimensional ultrasound to record the location of nodes,number,size,shape,border,internal echo,internal calcification, rear echo,acoustic halo,etc,conventional color Doppler to record distribution of blood ve ssels type in the nodule,resistance index and pulsation index.And then to ultrasound elasticity imaging,capture and storage elastic graph,complete classification on nodul es, at the same time record score and nodules of the elastic strain rate ratio.Finally ,a Logistic model on the basis of ultrasonographic features was obtained. Results: By Logistic regression analysis to evaluate the sonographic features of thyroid nod ule,three ultrasonic indexes including the boundary of the nodules、internal echo an d elasticity score are statistically significant (P<0.05).The areas under the ROC cu rve were 0.78、0.757 and 0.796(P<0.001).Conclusion:①The Logistic model can help differential diagnosis of benign and malignancy thyroid nodules. ②Most of t hyroid carcinoma shows fuzzy boundaries, internal hypoechoic and elasticity score 3 points or more.③Ultrasound elastography can objectively reflect the lesion tissu e stiffness information,for the identification of benign and malignant thyroid nodule s have certain clinical significance.