| 陈秋霞,刘奇,刘剑,彭玉兰,马步云.灰阶超声鉴别诊断良、恶性乳腺肿瘤[J].中国医学影像技术,2012,28(4):778~782 |
| 灰阶超声鉴别诊断良、恶性乳腺肿瘤 |
| Gray scale ultrasound in the differentiation of benign and malignant breast tumors |
| 投稿时间：2011-08-17 修订日期：2011-10-14 |
| DOI： |
| 中文关键词: 超声检查 乳腺肿瘤 形态特征 |
| 英文关键词:Ultrasonography Breast neoplasms Morphologic feature |
| 基金项目:西华大学四川省信号与信息处理重点实验室基金项目(SGXZD0101-10-1)。 |
|
| 摘要点击次数: 3154 |
| 全文下载次数: 1099 |
| 中文摘要: |
| 目的 探讨灰阶超声鉴别良、恶性乳腺肿瘤的价值。 方法 利用改进的Level Set变分模型对126例乳腺肿瘤的超声图像进行分割,提取肿瘤边界,分别计算16个形态特征参数,结合特征参数间的相关性及部分特征参数性质确定特征向量组合,最后用模糊C-均值方法鉴别乳腺肿瘤的良、恶性。 结果 126例中,恶性肿瘤50例,良性肿瘤76例。通过Level Set模型得到了较好的分割良、恶性的准确率达80.95%(102/126),其敏感度、特异度、阳性预测值和阴性预测值分别为80.00%(40/50)、81.58%(62/76)、74.07%(40/54)和86.11%(62/72)。 结论 良、恶性乳腺肿瘤在形态上有较大差异,灰阶超声可有效鉴别乳腺肿瘤的性质。 |
| 英文摘要: |
| Objective To assess the value of gray scale ultrasound in the differentiation of benign and malignant breast nodules. Methods Ultrasound images of 126 patients with breast nodules were segmented with the modified Level Set variational model. The contours of the breast nodules were extracted, and then 16 morphologic features were calculated. Combined with the correlation among those features and property of some features, the malignant and benign breast nodules were differentiated by means of fuzzy C-means algorithm. Results Among 126 breast nodules, there were 50 malignant and 76 benign. The segmentation result is quite good, and the diagnosis accuracy was up to 80.95% (102/126), the sensitivity, specificity, positive predictive value and negative predictive value was 80.00% (40/50), 81.58% (62/76), 74.07% (40/54) and 86.11% (62/72), respectively. Conclusion The benign and malignant breast nodules have significant variation in terms of the shape, and the gray scale ultrasound is effective in differentiating the malignant and benign breast nodules. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
|
|
|
