摘要: |
【】 目的:探讨乳腺癌患者超声影像组学定量特征分析与激素受体表达诊断相关性及其临床应用价值。方法:回顾性分析125例乳腺癌患者的术前超声特征与术后病理免疫组化结果。根据ER、PR、HER-2表达,将患者分为两组:激素受体阳性组(ER+、PR+、HER-2-),激素受体阴性组(ER-、PR-、HER-2-)。回顾性分析与评价乳腺癌肿块超声图像特征,通过基于相位信息动态轮廓模型筛选出与激素受体相关性强的特征性参数,应用支持向量机分类器及径向基核函数对研究数据进行研究分析。结果:超声定量组学代表性特征参数与激素受体表达相关联表明,除病灶形态外,两组间病灶大小、边缘、内部回声、后方回声、钙化等差异有统计学意义(P<0.05)。结论:超声影像组学定量特征分析可降低传统超声成像结论评价主观性,对乳腺癌激素受体表达特征及生物学行为预测方面具有一定的价值。 |
关键词: 乳腺癌 超声影像组学 定量特征 受体表达 |
DOI: |
投稿时间:2019-01-24修订日期:2019-02-27 |
基金项目: |
|
Clinical study on correlation between breast cancer hormone receptor expression diagnosis and ultrasound imaging omics |
LIU Biao-hu |
(Department of Ultrasound) |
Abstract: |
Objective To explore the correlation between the quantitative analysis of ultrasound imaging and the diagnosis of hormone receptor expression in breast cancer patients and its clinical application value.Methods The preoperative ultrasound features and postoperative pathological immunohistochemistry results of 125 patients with breast cancer were retrospectively analyzed. According to the expression of ER, PR and HER-2, patients were divided into two groups: hormone receptor positive group (ER+, PR+, HER-2-) and hormone receptor negative group (ER-, PR-, HER-2-). . Retrospectively analyze and evaluate the characteristics of ultrasound images of breast cancer masses, select the characteristic parameters with strong hormone receptors based on the dynamic contour model of phase information, and study the research data by using support vector machine classifier and radial basis kernel function. analysis.Results The correlation between the characteristic parameters of ultrasound quantitative group and the expression of hormone receptors showed that the lesion size, margin, internal echo, posterior echo and calcification were statistically significant (P<0.05). Conclusion Ultrasound imaging Quantitative Analysis can reduce the subjectivity of traditional ultrasound imaging evaluation, and has certain value for breast cancer hormone receptor expression and biological behavior prediction. |
Key words: Breast cancer Ultrasoundomics Quantitative characteristics Receptor expression |