| 李丽艳,周顺科,刘军,肖立志,唐中华,孙划.MSCT评价乳腺癌腋窝淋巴结转移[J].中国医学影像技术,2012,28(8):1529~1532 |
| MSCT评价乳腺癌腋窝淋巴结转移 |
| MSCT evaluation on axillary lymph nodes metastasis in patients with breast cancer |
| 投稿时间：2012-02-10 修订日期：2012-03-26 |
| DOI： |
| 中文关键词: 乳腺肿瘤 淋巴结 体层摄影术,X线计算机 判别分析 |
| 英文关键词:Breast neoplasms Lymph nodes Tomography, X-ray computed Discriminant analysis |
| 基金项目:中南大学创新基金项目(ZRD28)。 |
|
| 摘要点击次数: 3388 |
| 全文下载次数: 997 |
| 中文摘要: |
| 目的 探讨乳腺癌腋窝淋巴结转移的MSCT特征。方法 对23例乳腺癌患者行MSCT增强扫描,选取93枚最大径≥10 mm的腋窝淋巴结,观察腋窝淋巴结转移与其形态学特征的关系,同时与病理结果对照。通过Bayes逐步判别分析法筛选有价值的指标,建立腋窝淋巴结转移的判别诊断模型。结果 淋巴结最大径/最小径、淋巴结门、淋巴结实质、卫星淋巴结对诊断腋窝淋巴结转移有统计学意义。本组所建立的判别淋巴结转移的Bayes函数为:非转移淋巴结组:Y1=1.675X3(形态)+0.096X4(淋巴结门)+1.363X5(淋巴结实质)-0.612X7(卫星淋巴结)-1.033;转移淋巴结组:Y2=4.177X3(形态)+4.493X4(淋巴结门)+4.509X5(淋巴结实质)+2.351X7(卫星淋巴结)-6.490,经自身检证法与交互验证法考核判别诊断模型准确率分别达90.32%(84/93)、88.17%(82/93)。结论 用Bayes逐步判别分析方法筛选的MSCT征象指标及所建立的模型对诊断乳腺癌腋窝淋巴结转移有积极意义。 |
| 英文摘要: |
| Objective To investigate MSCT characteristics of axillary lymph nodes (ALNs) metastasis in patients with breast cancer. Methods Ninety-three ALNs with the longest axis ≥ 10 mm were detected by contrast-enhanced MSCT in 23 patients with breast cancer, and the results were compared with those of histopathologic examination. The correlation between metastasis of ALNs and its morphological features was analyzed. Bayes discriminant analysis was used to select the significant indexes, and the discriminant diagnostic cast and quantitative diagnostic schedule were established to diagnose ALNs metastasis in patients with breast cancer. Results The factors including the ratio of maximum diameter to minimum diameter, the hilum, the cortex and periphery changes of the ALNs showed statistical significance in diagnosis of ALNs metastasis in breast cancer. Two equations were constructed with Bayes discriminatory analysis for diagnosis of ALNs metastasis. The equation for non-metastatic ALNs was: Y1=1.675X3(shape)+0.096X4(hilum)+1.363X5(cortex)-0.612X7(satellite)-1.033, while for metastatic ALNs was: Y2=4.177X3(shape)+4.493X4(hilum)+4.509X5(cortex)+2.351X7(satellite)-6.490. The data of determined ALNs were resubstituted into the discriminatory function, and the results showed that the correct rate reached 90.32% (84/93). The accurate rate of diagnostic model was 88.17% (82/93) with cross-validated method. Conclusion Bayes discriminant analysis is of positive significance in differential diagnosis of ALNs through screening MSCT characteristics indexes of ALNs metastasis in patients with breast cancer. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
|
|
|
