Available value of semi-quantitative scoring system for contrast-enhanced ultrasound quantitative analysis's color images in the differential diagnosis of breast nodules
10.3760/cma.j.issn.1004-4477.2015.09.013
- VernacularTitle:超声造影定量分析彩色图像半定量评分系统在乳腺结节鉴别诊断中的价值
- Author:
Jun LUO
;
Jidong CHEN
;
Qing CHEN
;
Linxian YUE
;
Guo ZHOU
;
Cheng LAN
;
Yi LI
;
Chihua WU
;
Xuezhi SU
;
Jingqiao. LU
- Publication Type:Journal Article
- Keywords:
Ultrasonography;
Microbubbles;
Breast diseases;
Diagnosis,differenital
- From:
Chinese Journal of Ultrasonography
2015;(9):784-788
- CountryChina
- Language:Chinese
-
Abstract:
Objective To assess the feasibility of semi-quantitative scoring system for contrast-enhanced ultrasound (CEUS)quantitative analysis's color images in the differential diagnosis of breast nodules.Methods Totally 244 BI-RADS 4 breast solid lesions received CEUS before core needle biopsy or surgical resection were included.A semi-quantitative scoring system for color images of CEUS quantitative analysis were built.The scores were given as follows:1 )Color type and its distribution (0 to 4);2)Color scope (0 to 1 );3)Color margin (0 to 1 );4)Color shape (0 to1 ).The total score for each lesion would be from 0 to 7.And the differenital value between benign and malignant lesions were assessed.Results The total semi-quantitative scores of 102 malignant tumors (5.1 ±1 .7)was significant higher than that of benign lesions (3.34±0.7)(P < 0.05 ).In 102 malignant lesions,the total scores of 81 lesions (79.41 %)were more than 4 points,and in 142 benign lesions,the total scores of 89 lesions (62.67%)were less than 4 points.Depending on the Wilcox rank sum test (Mann-Whitney)analysis,the distribution of total scores between benign and malignant lesions was significant different (P <0.000 1).Total score 4 was selected as the best cutoff,the area under ROC curve was 0.749,on which the sensitivity,specificity and accuracy were 79.4%,62.7% and 69.67%,respectively.Conclusions The semi-quantitative scoring system of CEUS quantitative analysis color images showed good sensitivity but not satisfied specificity and accuracy in differential diagnosis between malignant and benign breast lesions.
