| 陈艳,韩义娜,郭亚飞,吕以东,谭红娜,赵鑫.乳腺X线片直方图分析鉴别乳腺BI-RADS 3~5类良恶性肿块[J].中国医学影像技术,2020,36(11):1648~1651 |
| 乳腺X线片直方图分析鉴别乳腺BI-RADS 3~5类良恶性肿块 |
| Differential diagnosis of benign and malignant BI-RADS category 3-5 breast masses with histogram analysis of breast X-ray images |
| 投稿时间：2019-10-11 修订日期：2020-06-18 |
| DOI：10.13929/j.issn.1003-3289.2020.11.013 |
| 中文关键词: 乳腺肿瘤 乳房X线摄影术 直方图分析 |
| 英文关键词:breast neoplasms mammography histogram analysis |
| 基金项目:中国博士后科学基金（2018M632779）。 |
|
| 摘要点击次数: 2396 |
| 全文下载次数: 684 |
| 中文摘要: |
| 目的 探讨基于乳腺X线片直方图分析鉴别乳腺影像报告和数据系统（BI-RADS）3~5类良恶性肿块的价值。方法 回顾性分析经手术病理证实的114例BI-RADS 3~5类乳腺肿块患者，包括61例良性病变（良性组，68个肿块）和53例恶性病变（恶性组，55个肿块）。分析2组图像的直方图，比较组间直方图参数差异，包括平均值、方差、偏斜度、峰度及第1、10、50、90、99百分位数；分别绘制差异有统计学意义的参数鉴别诊断BI-RADS 3~5类乳腺良恶性肿块的受试者工作特征（ROC）曲线，计算曲线下面积（AUC），分析其诊断效能。结果 2组平均值、方差及第1、10、50、90、99百分位数差异均有统计学意义（t=－5.49、－3.14、－3.01、－3.97、－5.49、－5.84、－6.45，P均<0.05）。ROC曲线分析结果显示第99百分位数诊断效能最佳，其AUC为0.81，最佳阈值为0.50时，特异度为88.20%，敏感度为61.80%。结论 乳腺X线片直方图分析可用于鉴别诊断乳腺BI-RADS 3~5类良恶性肿块。 |
| 英文摘要: |
| Objective To explore the value of histogram analysis of breast X-ray images for differentiating of breast imaging reporting and data system (BI-RADS) category 3-5 benign and malignant breast masses. Methods A retrospective analysis was made from 114 patients with BI-RADS category 3-5 breast masses confirmed by pathology, including 61 patients with 68 benign lesions (benign group) and 53 with 55 malignant lesions (malignant group). Histogram analysis was performed on images of the two groups, and the histogram parameters, including average value, variance, skewness, kurtosis, as well as the 1st, 10th, 50th, 90th and 99th percentiles were compared between 2 groups. Receiver operating characteristic (ROC) curves of parameters being statistically different between groups were drawn for differentiating benign and malignant BI-RADS 3-5 breast masses, respectively, and the corresponding area under the curve (AUC) and diagnostic efficacy were measured and analyzed. Results Statistical significances of average, variance, 1st, 10th, 50th, 90th and 99th percentiles were found between 2 groups (t=－5.49, －3.14, －3.01, －3.97, －5.49, －5.84, －6.45, all P<0.05). ROC curve analysis showed that the 99th percentile had the best diagnostic performance, with AUC of 0.81, and taken the optimal threshold as 0.50, the specificity and sensitivity was 88.20% and 61.80%, respectively. Conclusion Histogram analysis based on breast X-ray images could be used to differentiating BI-RADS 3-5 benign and malignant breast masses. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
|
|
|
