基于特征矢量化的肺结节特征选择算法

被引:3
作者
贺兴怡
龚敬
王丽嘉
聂生东
机构
[1] 上海理工大学医学影像工程研究所
基金
上海市自然科学基金;
关键词
特征选择; 肺结节; Relief算法; 计算机断层扫描;
D O I
暂无
中图分类号
R734.2 [肺肿瘤]; TP391.41 [];
学科分类号
100117 [系统生物医学];
摘要
针对肺结节良/恶性分类模型中特征选择过程无法避免特征多样性不受破坏的问题,提出一种将肺结节特征矢量化处理的特征选择方法。假设每个肺结节特征都是由数据、类型构成的一个矢量,按照特征类型添加特征到相应的特征子集,并分别利用Relief算法评价特征、特征子集的分类重要性。通过动态阈值的方式筛选得到优化后的特征子集。在150个肺结节样本的分类实验中,采用提出的算法所取得的敏感性为94.7%、特异性为93.7%、虚警率为5.2%、受试者工作特性曲线下面积为97.3%。分析表明,提出的算法几乎不破坏肺结节特征的多样性,能够显著提高肺结节良/恶性分类的准确性。
引用
收藏
页码:2544 / 2548
页数:5
相关论文
共 22 条
[1]
孤立性肺结节良恶性综合性影像诊断的ROC分析及其临床应用价值研究 [D]. 
杨玉海 .
山东大学,
2011
[2]
Differentiation of Benign and Malignant Solitary Pulmonary Nodule: Literature Review.[J].Suresh Tripathi;Xuqiu Zhen.Advances in Lung Cancer.2015, 2
[3]
Texture Feature Analysis for Computer-Aided Diagnosis on Pulmonary Nodules [J].
Han, Fangfang ;
Wang, Huafeng ;
Zhang, Guopeng ;
Han, Hao ;
Song, Bowen ;
Li, Lihong ;
Moore, William ;
Lu, Hongbing ;
Zhao, Hong ;
Liang, Zhengrong .
JOURNAL OF DIGITAL IMAGING, 2015, 28 (01) :99-115
[4]
Performance comparison of artificial neural network and logistic regression model for differentiating lung nodules on CT scans [J].
Chen, Hui ;
Zhang, Jing ;
Xu, Yan ;
Chen, Budong ;
Zhang, Kuan .
EXPERT SYSTEMS WITH APPLICATIONS, 2012, 39 (13) :11503-11509
[5]
Mapping LIDC; RadLex; and lung nodule image features..[J].Opulencia P;Channin DS;Raicu DS;Furst JD.Journal of digital imaging: the official journal of the Society for Computer Applications in Radiology.2011, 2
[6]
CAD (computed-aided detection) and CADx (computer aided diagnosis) systems in identifying and characterising lung nodules on chest CT: overview of research, developments and new prospects [J].
Fraioli, F. ;
Serra, G. ;
Passariello, R. .
RADIOLOGIA MEDICA, 2010, 115 (03) :385-402
[7]
A Novel Approach to Nodule Feature Optimization on Thin Section Thoracic CT [J].
Samala, Ravi ;
Moreno, Wilfrido ;
You, Yuncheng ;
Qian, Wei .
ACADEMIC RADIOLOGY, 2009, 16 (04) :418-427
[8]
Clinical tests: sensitivity and specificity [J].
Lalkhen, Abdul Ghaaliq ;
McCluskey, Anthony .
BJA EDUCATION, 2008, 8 (06) :221-223
[9]
Receiver-operating characteristic analysis for evaluating diagnostic tests and predictive models [J].
Zou, Kelly H. ;
O'Malley, A. James ;
Mauri, Laura .
CIRCULATION, 2007, 115 (05) :654-657
[10]
Thirteen Ways to Look at the Correlation Coefficient.[J].Rodgers Joseph Lee;Nicewander W. Alan.The American Statistician.1988, 1