The value of ultrasonic images texture analysis in differential diagnosis of parotid pleomorphic adenoma and adenolymphoma#br#
HAN Wen1, LI Weimin2, LING Li1, CHEN Wei1
(1. Department of Ultrasound, People′s Hospital of Suzhou National New Hitech Industrial, Suzhou Jiangsu 215129; 2. Department of Ultrasound, Affiliated Hospital of Jiangnan University, Wuxi Jiangsu 214000, China)
Abstract:Objective: To investigate the value of ultrasonic images texture analysis in differential diagnosis of parotid pleomorphic adenoma and adenolymphoma. Methods: A total of 43 parotid pleomorphic adenomas and 54 adenolymphomas were selected. The ultrasonic images were imported into Mazda 4.6 software and ROIs were manually drawed. The optimum texture parameters were selected through mutual information (MI), Fisher coefficient, probability of classification error and average correction coefficient (POE+ACC) and the combination of three methods (Fisher+POE+ACC+MI), respectively.The model of artificial neural network was built. The assessment results of texture analysis and ultrasonic doctors were compared with the pathology. And misdiagnostic rates were analyzed between texture analysis and ultrasonic doctors. Results: The age and proportion of male in patients with parotid adenolymphoma were higher than those with parotid pleomorphic adenoma,there were statistically difference (P<0.001). In ultrasonic features, there were statistical difference between cystic degeneration and blood flow in two groups (P<0.001 or P<0.05).Among the 30 groups of texture parameters extracted by Mazda, there were 8 groups with statistical differences (all P<0.05).The misdiagnostic rates of MI, POE+ ACC, Fisher and Fisher+POE+ ACC+MI were 53.61%(52/97), 44.32%(43/97), 21.65%(21/97), 17.53%(17/97), respectively. Furthermore, the misjudgment rate of the combination of three methods was lower than that of the ultrasound doctors(29.90%, 29/97), there was statistical differences(χ2=4.103, P<0.05). Conclusion:Ultrasonic images texture analysis can be used to differentiate parotid pleomorphic adenoma and adenolymphoma.
韩文,李卫民,凌莉,等.. 超声影像纹理分析鉴别诊断腮腺混合瘤和腺淋巴瘤的价值[J]. 江苏大学学报:医学版, 2022, 32(05): 409-414.
HAN Wen1, LI Weimin2, LING Li1, CHEN Wei1. The value of ultrasonic images texture analysis in differential diagnosis of parotid pleomorphic adenoma and adenolymphoma#br#. Journal of Jiangsu University(Medicine Edition), 2022, 32(05): 409-414.
[1]Mantsopoulos K, Iro H. Pleomorphic adenoma compared with cystadenolymphoma of the parotid gland: which is more common?[J]. Br J Oral Maxillofac Surg, 2020, 58(3): 361-363.
[2]黄冠兰,徐秋华,燕山. 腮腺多形性腺瘤与腺淋巴瘤的超声鉴别研究[J]. 中华超声影像学杂志, 2013, 22(4): 366-367.
[3]江晓勇,杨志辉,陈希希,等. 不同类型腮腺肿瘤的临床特点及远期复发情况分析[J]. 中华全科医学, 2019, 17(5): 790-792, 796.
[4]方三高. WHO(2017)头颈部肿瘤分类[J]. 诊断病理学杂志, 2017, 24(8): 638-640.
[5]Thomas R, Qin L, Alessandrino F, et al. A review of the principles of texture analysis and its role in imaging of genitourinary neoplasms[J]. Abdom Radiol(NY), 2019, 44(7): 2501-2510.
[6]岳茜,李俊,张莉,等. CT纹理分析在消化系统恶性肿瘤中的应用进展[J]. 中国医学影像技术, 2019, 35(3): 136-139.
[7]Skogen K, Schulz A, Dormagen JB, et al. Diagnostic performance of texture analysis on MRI in grading cerebral gliomas[J]. Eur J Radiol, 2016, 85(4): 824-829.
[8]刘文华, 张衡, 李敏,等. CT图像纹理分析鉴别诊断腮腺混合瘤与腺淋巴瘤[J]. 临床放射学杂志, 2019, 38(12): 2271-2274.
[9]宋晶,刘丽,王蕾,等.高频超声在腮腺腺淋巴瘤和腮腺混合瘤的鉴别诊断研究[J].中国医学装备, 2019, 16(7): 109-111.
[10]韩明丽,王栋华,史秋生,等.基于超声特征的评分法在腮腺Warthin瘤与混合瘤鉴别诊断中的应用研究[J].中国临床医学影像杂志, 2020, 31(6): 385-389.
[11]Kanatas A, Ho MWS, Mücke T. Current thinking about the management of recurrent pleomorphic adenoma of the parotid: a structured review[J]. Br J Oral Maxillofac Surg, 2018, 56(4): 243-248.
[12]Rozenberg R, Thornhill RE, Flood TA, et al. Wholetumor quantitative apparent diffusion coefficient histogram and texture analysis to predict gleason score upgrading in intermediaterisk 3+4=7 prostate cancer[J]. AJR Am J Roentgenol, 2016, 206(4): 775-782.
[13]Canellas R, Burk KS, Parakh A, et al. Prediction of pancreatic neuroendocrine tumor grade based on CT features and texture analysis[J]. AJR Am J Roentgenol, 2018, 210(2): 341-349.
[14]张竹伟, 华婷, 徐婷婷,等. 常规MRI纹理分析鉴别乳腺良、恶性病变的价值初探[J]. 中华放射学杂志, 2017, 51(8): 588-591.