Clinical application of S-Detect on breast masses on ultrasonography: a study analyzing the influence factors and #br#
the combined diagnosis with a radiologist specializing in breast lesions
YAN Hong, LI Xiang, CHENG Hui-fang, WANG Xue-mei, ZHANG Yi-xia, KANG Shu
Department of Cardiovascular Ultrasound, the First Hospital of China Medical University, Shenyang 110001, China
Abstract:Objective: To investigate the influence factors of the sensitivity, specificity and accuracy of S-Detect. To analyze the diagnostic value and the agreement of the combined diagnosis with a radiologist specializing in breast lesions. Methods: Grouped by the influence factors of S-Detect, chi-square test was used to calculate and compare the sensitivity, specificity and accuracy of S-Detect. Logistic regression was used to analyze the relationship between various factors and the results of pathological diagnosis that was benign or malignant. Breast ultrasound and S-Detect were operated by a radiologist specializing in breast imaging. The other radiologist analyzed the ultrasonographic images of the breast masses according to the American College of Radiology Breast Imaging Reporting and Data System(BI-RADS), and then made the combined diagnosis on the results of S-Detect. The area under the receiver operating characteristic(ROC) curve and the agreement among the S-Detect, radiologist, combined diagnosis were calculated and compared. Results: In all the 581 lesions, pathological diagnosis showed 411 benign lesions and 170 malignant lesions. The false positive rate of S-Detect in intraductal papilloma and which was between 2 cm to 4 cm was higher(P<0.05). The specificity of S-Detect was all higher than 80.00%. The sensitivity and specificity of S-Detect were not statistically significant in the groups of the age, the location of the lesions and the depth of the lesion. Three significant risk factors(age, S-Detect, the most largest size of lesions) for pathology were analyzed with logistic regression(P<0.05). Moderate agreement was seen in final assessments made by S-Detect, radiologist and combined diagnosis(the primary combined diagnosis)(Kappa=0.45, 0.571 and 0.65, respectively) with pathological results as the gold standard. Good agreement was seen in final assessments made by combined diagnosis in consideration of age and the largest size of lesions(the recombined diagnosis)(Kappa=0.76). The area under ROC curve was 0.78, 0.78, 0.84 and 0.89, respectively. Conclusion: The specificity of S-Detect was higher. The accuracy of S-Detect in intraductal papilloma was lower and the false negative rate of S-Detect in benign tumor of lesions between 2 cm to 4 cm was higher. S-Detect is a clinically feasible diagnostic tool that can be used to improve the diagnostic value and the agreement of the radiologist and pathological diagnosis.
闫 虹,李 响,程慧芳,王学梅,张义侠,康 姝. S-Detect技术应用于超声诊断乳腺包块的影响因素及与超声医师联合诊断的分析[J]. 中国临床医学影像杂志, 2020, 31(1): 24-29.
YAN Hong, LI Xiang, CHENG Hui-fang, WANG Xue-mei, ZHANG Yi-xia, KANG Shu. Clinical application of S-Detect on breast masses on ultrasonography: a study analyzing the influence factors and #br#
the combined diagnosis with a radiologist specializing in breast lesions. JOURNAL OF CHINA MEDICAL IMAGING, 2020, 31(1): 24-29.
[1]彭卫军,顾雅佳. 重视乳腺影像学检查技术和诊断水平的普及及提高为降低我国乳腺癌发病率而努力[J].中华放射学杂志,2009,43(5):453-454.
[2]Au FW, Ghai S, Moshonov H, et al. Diagnostic performance of quantitative shear wave elastography in the evaluation of solid breast masses: determination of the most discriminatory parameter[J]. Am J Roentgenol, 2014, 203(9): W328-W336.
[3]Kim EJ, Kim SH, Kang BJ, et al. Interobserver agreement on the interpretation of automated whole breast ultrasonography[J]. Ultrasonography, 2014, 33(4): 252-258.
[4]Shin HJ, Kim HH, Cha JH, et al. Automated ultrasound of the breast for diagnosis: interobserver agreement on lesion detection and characterization[J]. Am J Roentgenol, 2011, 197(3): 747-754.
[5]Chang RF, Wu WJ, Moon WK, et al. Automatic ultrasound segmentation and morphology based diagnosis of solid breast tumors[J]. Breast Cancer Res Treat, 2005, 89(2): 179-185.
[6]Chen CM, Chou YH, Han KC, et al. Breast lesions on sonograms: computer-aided diagnosis with nearly setting-independent features and artificial neural networks[J]. Radiology, 2003, 226(2): 504-514.
[7]Drukker K, Giger ML, Vyborny CJ, et al. Computerized detection and classification of cancer on breast ultrasound[J]. Acad Radiol, 2004, 11(5): 526-535.
[8]Kim KG, Cho SW, Min SJ, et al. Computerized scheme for assessing ultrasonographic features of breast masses[J]. Acad Radiol, 2005, 12(1): 58-66.
[9]Sehgal CM, Cary TW, Kangas SA, et al. Computer-based margin analysis of breast sonography for differentiating malignant and benign masses[J]. J Ultrasound Med, 2004, 23(9): 1201-1209.
[10]Kim K, Song MK, Kim EK, et al. Clinical application of S-Detect to breast masses on ultrasonography: a study evaluating the diagnostic performance and agreement with a dedicated breast radiologist[J]. Ultrasonography, 2017, 36(1): 3-9.
[11]王洁贞,韩兢,刘言训,等. Kappa 统计量在一致性和重现性检验中的应用[J]. 山东医科大学学报,1996,34(3):209-212.
[12]姜海艳,刘爱玲,韩东明. S-Detect TM技术在乳腺肿物诊断中的应用分析[J]. 河南医学研究,2016,25(6):979-981.
[13]Choi JH, Kang BJ, Baek JE, et al. Application of computer-aided diagnosis in breast ultrasound interpretation: improvements in diagnostic performance according to reader experience[J]. Ultrasonography, 2018, 37(3): 217-225.
[14]Cho E, Kim EK, Song MK, et al. Application of Computer-Aided Diagnosis on Breast Ultrasonography: Evaluation of Diagnostic Performances and Agreement of Radiologists According to Different Levels of Experience[J]. J Ultrasound Med, 2017, 37(1): 209-216.
[15]Di MS, De VS, Cantisani V, et al. Automated classification of focal breast lesions according to S-detect: validation and role as a clinical and teaching tool[J]. J Ultrasound, 2018, 21(2): 105-118
[16]王立娟,勇强,包晶晶,等. 应用剪切波技术测量模块弹性值符合度的试验研究[J]. 中国超声医学杂志,2018,34(7):655-659.
[17]苏彤,邓晶,龚海燕,等. 弹性成像参数对乳腺肿块的诊断价值[J]. 肿瘤影像学,2019,28(2):90-95.
[18]苏晖,刘翠云,陈烨,等. 超声自动全乳腺容积扫描(ABVS)与常规二维超声在乳腺肿块良恶性对比诊断中的应用价值[J]. 心电图杂志:电子版,2018,7(4):119-120.
[19]Zeiler MD, Fergus R. Visualizing and understanding convolutional networks. In: Proc. of the European Conf. on Computer Vision[C]. Zurich: Springer-Verlag, 2014: 818-833.