超声BI-RADS分类联合弹性成像在乳腺肿块良恶性诊断中的应用价值

发布时间：2018-04-08 10:36

本文选题：BI-RADS分类　切入点：乳腺肿块　出处：《兰州大学》2017年硕士论文

【摘要】：目的探讨2013版BI-RADS分类、单独使用不同超声弹性评估方法(2013版BI-RADS超声弹性分级法、弹性应变率比值法、面积比值法)以及两者联合的方法在乳腺肿块良恶性鉴别中的诊断价值。方法选取98例患者,共计103个肿块,所有肿块均用常规超声和弹性成像两种方式进行检查,按照2013版BI-RADS分类标准进行分类,所有病灶行穿刺活检术或手术切除术,获取标本送病理诊断,对照病理结果构建ROC曲线确定两种弹性评估方法(SR法和AR法)的最佳诊断截断点,用三种不同的弹性评估方法对肿块的性质进行评判,再根据不同的弹性评估结果对BI-RADS分类进行调整,比较不同方法之间的差异。结果1. BI-RADS分类在进行乳腺肿块良恶性诊断时的灵敏度是93. 85%,特异度是94. 74%,准确度是 94. 17%。2.弹性分级法、应变率比值法、面积比值法诊断乳腺肿块良恶性的准确度分别为 70. 87%,82. 52%,82. 52%;灵敏度分别为 90. 24%,90. 91%,90. 91%;特异度分别为 58. 06%,72. 92%,72. 92%。3. BI-RADS分类联合三种不同弹性评估方法(分级法、SR法、AR法)诊断乳腺肿块良恶性的准确度分别为91. 26%,90.29%,89. 32%;灵敏度分别为93. 55%,93. 44%, 91.94%;特异度分别为87. 8%,85. 71%,85. 37%。4.构建ROC曲线,BI-RADS分类、不同弹性评估方法(分级法、SR法、AR法)、BI-RADS分类联合三种不同弹性评估方法(分级法、SR法、AR法)ROC曲线下的面移积分别为0. 952, 0. 778, 0. 843, 0. 892, 0. 951,0. 965, 0. 952。结论1. 2013版BI-RADS分类、单独使用不同的弹性成像评估方法、两种方法联合在乳腺病灶良恶性鉴别中均具有诊断价值。2.单独使用不同的弹性成像评估方法进行诊断时,超声弹性分级法有一定的局限性,应变率比值法、面积比值法作为一种半定量评价方法在一定程度可以避免医师的主观影响比其具有更高的诊断价值。3.二维超声是诊断的基础,弹性成像技术可以作为常规超声的补充,为BI-RADS分类提供更多的参考信息,提升超声医生的诊断信心,为临床诊疗提供更好的服务。
[Abstract]:Objective to explore the diagnostic value of BI-RADS classification in 2013 edition, using different ultrasonic elastic assessment methods alone, and using the different ultrasound elastic classification method, elastic strain rate ratio method, area ratio method and combined method in the differential diagnosis of benign and malignant breast masses.Methods 98 patients with 103 masses were selected. All the masses were examined by conventional ultrasound and elastic imaging. The lesions were classified according to the 2013 BI-RADS classification standard, and all lesions were performed puncture biopsy or surgical excision.ROC curve was constructed to determine the best diagnostic cutoff point of two elastic evaluation methods: SR method and AR method. Three different elastic evaluation methods were used to evaluate the nature of the tumor.Then the BI-RADS classification is adjusted according to different elastic evaluation results, and the differences between different methods are compared.Result 1.The sensitivity of BI-RADS classification in the diagnosis of benign and malignant breast masses was 93. 3%.The specificity is 94.The accuracy is 94.17. 2.The accuracy of elastic grading method, strain rate ratio method and area ratio method in the diagnosis of benign and malignant breast masses were 70. 0% respectively.87s 82s.52 / 82.The sensitivity is 90.Two hundred and ninety.91s and 90s.The specificity was 58.06and 72s.Two hundred and seventy-two.Nine two and three.The accuracy of BI-RADS classification combined with three different elastic assessment methods (classification / SR / AR) in the diagnosis of benign and malignant breast masses was 91. 1% respectively.There are 90.29 and 89.The sensitivity is 93. 3% respectively.Five hundred and fifty three.The specificity was 87. 4% and 91. 94% respectively.Eight hundred and eighty-five.71s and 85s.370.4.BI-RADS classification of ROC curve was constructed. Different elastic evaluation methods (classification method / SR method / AR method / AR method) combined with three different elastic assessment methods (classification method / SR method / AR method / surface shift product under ROC curve were 0 respectively).952, 0.778, 0.843, 0.892, 0.951,0.965, 0.952.Conclusion 1.2013 version of BI-RADS classification, using different methods of elastography alone, the two methods combined in the diagnosis of benign and malignant breast lesions are of value. 2.The ultrasonic elastic classification method has some limitations when it is used in the diagnosis of different elastic imaging evaluation methods alone, the strain rate ratio method, the strain rate ratio method, the strain rate ratio method, the strain rate ratio method,As a semi-quantitative evaluation method, the area ratio method can avoid the subjective influence of doctors to a certain extent.Two-dimensional ultrasound is the basis of diagnosis, elastic imaging technology can be used as a supplement to conventional ultrasound, provide more reference information for BI-RADS classification, enhance the diagnosis confidence of ultrasound doctors, and provide a better service for clinical diagnosis and treatment.
【学位授予单位】：兰州大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：R445.1;R737.9

【参考文献】