基于邻域关系模糊粗糙集的医学图像分类研究

摘要/Abstract

摘要：

对医学图像进行分类时，特征选择是影响分类准确率的非常重要的因素。针对医学图像的特殊性，以及目前提出的特征选择算法在应用于医学图像分类时效果不够理想等问题，提出一种基于邻域关系的模糊粗糙集模型，基于该模型给出特征选择算法，并将其应用于乳腺X光图像。实验结果表明，同已有的算法相比，该方法能有效选择特征，分类精度有较大的提升。

关键词: 医学图像分类, 特征选择, 邻域关系

Abstract:

For medical images classification, feature selection is an important factor that affects classification accuracy. Aiming at the particularity of medical image as well as undesirable effects of current feature selection methods, we propose a fuzzy rough set model based on neighborhood relation, and a feature selection algorithm based on the model as well. We apply our algorithm to mammography, and the experimental results show that it can effectively select features and improve classification accuracy in comparison with the existing algorithms.

Key words: medical image classification;feature selection;neighborhood relation

胡学伟，蒋芸，邹丽，李志磊，沈健. 基于邻域关系模糊粗糙集的医学图像分类研究[J]. J4, 2016, 38(04): 739-746.

HU Xuewei,JIANG Yun,ZOU Li,LI Zhilei,SHEN Jian. Medical image classification based on
neighborhood relation fuzzy rough set [J]. J4, 2016, 38(04): 739-746.

参考文献［28］

［1］	Nishikawa R M. Current status and future directions of computeraided diagnosis in mammography［J］. Computerized Medical Imaging and Graphics,2007,31:224235.
［2］	Duncan J S,Ayache N. Medical image analysis:Progress over two decades and the challenge ahead［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence,2000,22(1):85106.
［3］	Zhang Chao,Jiang Hongchuan. Application of Shuyi breast diagnostic instrument in breast cancer diagnosis［J］. Chinese Journal of Cancer Prevention and Treatment,2010,17(19):16001604.(in Chinese)
［4］	Xie Guocheng. Research on medical image classification method based on hypersphere multiclass support vector data description［D］. Lanzhou:Northwest Normal University,2014. (in Chinese)
［5］	Yue Lijuan. Research on medical image feature extraction and classification ［D］. Wuxi:Jiangnan University,2013.(in Chinese)
［6］	Xu Jiucheng, Xu Tianhe, Sun Lin. Feature selection for cancer classification based on neighborhood rough set and particle swarm optimization［J］.Journal of Chinese Computer Systems,2014,35(11):25282532.(in Chinese)
［7］	Li Weiwei. The research of feature selection and segmentation of medical image based on fusion［D］.Suzhou:Suzhou University,2010.(in Chinese)
［8］	Yang Hongwei. Research and system implement on feature extraction & feature selection for pulmonary nodules［D］.Chongqing:Chongqing University,2010.(in Chinese)
［9］	Pawlak Z. Rough sets［J］.International Journal of Information and Computer Science,1982,11(5):129141.
［10］	Yao Yiyou. Threeway decisions with probabilistic rough sets［J］. Information Sciences,2010,180(3):341353.
［11］	Yao Yiyou. The superiority of threeway decisions in probabilistic rough set models［J］.Information Sciences,2011,18(6):10801096.
［12］	Yao Yiyou. Two semantic issues in a probabilistic rough set model［J］. Fundmental Informatic,2011,108(3/4):249265.
［13］	Sharma R, Jain P, Shrivastava S K. An optimize decision tree algorithm based on variable precision rough set theory using degree of βdependency and significance of attributes［J］. International Journal of Computer Science and Information Technologies,2012,3(3):39423947.
［14］	Zhang Wenxiu,Wu Weizhi,Liang Jiye. Rough set theory and method［M］. Beijing:Science Press,2005.(in Chinese)
［15］	Wang G Y,Ma X,Yu H. Monotonic uncertainty measures for attribute reduction in probabilistic rough set model［J］. International Journal of Approximate Reasoning,2015,59(3):4167.
［16］	Jia X,Tang Z,Liao W,et al. On an optimization representation of decisiontheoretic rough set model［J］. International Journal of Approximate Reasoning,2014,55(1):156166.
［17］	Nguyen X T,Nguyen D D. Rough fuzzy relation on two universal sets［J］. International Journal of Intelligent Systems and Applications (IJISA),2014,6(4):49.
［18］	Chen D,Yang Y. Attribute reduction for heterogeneous data based on the combination of classical and fuzzy rough set models［J］. IEEE Transactions on Fuzzy Systems,2014,22(5):13251334.
［19］	Lin T Y,Huang K J,Liu Q,et al. Rough sets,neighborhood systems and approximation［C］∥Proc of the 5th International Symposium on Methodologies for Intelligent System, 1990:130141.
［20］	Lingras P,Chen M,Miao D. Qualitative and quantitative combinations of crisp and rough clustering schemes using dominance relations［J］. International Journal of Approximate Reasoning,2014,55(1):238258.
［21］	Hu Qinghua,Yu Daren. Neighborhood rough set based heterogeneous feature subset selection［J］.Information Science,2008,178(18):35773594.
［22］	Hu Qinghua,Yu Daren,Xie Zongxia. Numerical attribute reduction based on neighborhood granulation and rough approximation［J］. Journal of Software,2008,19(3):640649.(in Chinese)
［23］	Hu Qinghua,Yu Daren,Xie Zongxia. Neighborhood classifiers［J］. Expert Systems with Applications,2008,34(2):866876.
［24］	The mammography image analysis society［DB/OL］.［20060901］.http://www.wiau.man.ac.uk/services/MIAS/MIAS web.html.
［25］	Ruan Qiuqi. Digital image processing［M］. Beijing:Publishing House of Electronics Industry,2001:481526.(in Chinese)
［26］	Youssry N,Abouchadi F E Z,ElSayad A M. Early detection of masses in digitized mammograms using texture features and neuron fuzz model［C］∥Proc of the 4th International IEEE EMBS Special Topic Conference on Information Technology Applications in Biomedicine,2003:226229.
［27］	Miao Duoqian,Li Daoguo. Rough sets theory algorithms and applications［M］. Beijing:Tsinghua University Press,2008:246247.(in Chinese)
［28］	Hu Qinghua,Yu Daren. Application of rough calculation［M］.Beijing:Science Press,2012:8081.(in Chinese)
［29］	Hu Qinghua,An Shuang,Yu Daren. Soft fuzzy rough sets for robust feature evaluation and selection［J］. Information Sciences,2010,180(22):43844400.
［30］	Kleefeld A,Breuβ M,Welk M,et al. Adaptive filters for color images:Median filtering and its extensions［C］∥Proc of the 5th International Computational Color Imaging Workshop(CCIW’05),2015:149158.
［31］	Liu Yong,Huang Wenliang,Jiang Yunliang,et al. Quick attribute reduct algorithm for neighborhood rough set model［J］. Information Sciences,2014,271(7):6581.
	附中文参考文献：
［3］	张超,蒋宏传. 舒怡乳腺诊断仪在乳腺癌诊断中的应用［J］. 中华肿瘤防治杂志,2010,17(19):16001604.
［4］	谢国城. 基于超球体多类支持向量数据描述的医学图像分类方法研究［D］. 兰州:西北师范大学,2014.
［5］	岳丽娟. 医学图像的特征提取与分类方法研究［D］. 无锡:江南大学,2013.
［6］	徐久成,徐天贺,孙林. 基于邻域粗糙集和粒子群优化的肿瘤分类特征基因选取［J］. 小型微型计算机系统,2014,35(11):25282532.
［7］	李卫伟. 基于融合的医学图像特征选择和分割方法研究［D］. 苏州:苏州大学,2010.
［8］	杨宏薇. 肺结节特征提取和特征选择的研究及系统实现［D］. 重庆:重庆大学,2010.
［14］	张文修,吴伟志,梁吉业.粗糙集理论与方法［M］.北京:科学出版社,2005.
［22］	胡清华,于达仁,谢宗霞. 基于邻域粒化和粗糙逼近的数值属性约简［J］.软件学报,2008,19(3):640649.
［25］	阮秋奇.数字图像处理学［M］.北京:电子工业出版社,2001:481526.
［27］	苗夺谦,李道国.粗糙集理论、算法与应用［M］.北京:清华大学出版,2008:246247.
［28］	胡清华,于达仁.应用粗糙计算［M］.北京:科学出版社,2012:8081.

[1]	赵瑞平, 降爱莲. 基于自编码器和局部嵌入的无监督特征选择[J]. 计算机工程与科学, 2023, 45(07): 1282-1291.
[2]	顾楚梅, 曹建军, 王保卫, 徐雨芯, . 基于蚁群参数优化的LightGBM辐射源个体识别[J]. 计算机工程与科学, 2023, 45(01): 85-94.
[3]	文武, 万玉辉, 文志云, . 基于正余弦算法的文本特征选择[J]. 计算机工程与科学, 2022, 44(08): 1467-1473.
[4]	刘云, 肖添, 王梓宇. 动态特征选择算法对恶意行为检测的优化研究[J]. 计算机工程与科学, 2022, 44(04): 665-673.
[5]	吴尚智, 徐丹丹, 王旭文, 夏宁. 基于广义重要度和runner-root算法的特征选择[J]. 计算机工程与科学, 2022, 44(04): 723-729.
[6]	李雨晨, 魏巍, 白伟明, 王达. 基于标签共现关系的多标签特征选择[J]. 计算机工程与科学, 2021, 43(11): 2049-2055.
[7]	文武, 万玉辉, 张许红, 文志云, . 基于改进CHI和PCA的文本特征选择[J]. 计算机工程与科学, 2021, 43(09): 1645-1652.
[8]	李向军1,2，孔珂2，魏智翔1，王科选1，肖聚鑫1. 面向Android恶意应用静态检测的特征频数差异增强算法[J]. 计算机工程与科学, 2020, 42(06): 993-1002.
[9]	孟昱煜，陈绍立，刘兴长. 面向排序学习的层次聚类特征选择算法[J]. 计算机工程与科学, 2019, 41(12): 2211-2216.
[10]	张守宾，朱习军. 集成学习算法在中医证型分类预测中的应用[J]. 计算机工程与科学, 2019, 41(02): 328-334.
[11]	纪明君，刘漫丹，才乐千. 基于半监督LDA特征子空间优化的人脸识别算法[J]. 计算机工程与科学, 2018, 40(10): 1851-1857.
[12]	吴帅，赵方. 基于随机森林的老年人居住偏好预测研究[J]. 计算机工程与科学, 2018, 40(05): 924-930.
[13]	杨超，蔡晓东，甘凯今，王丽娟. 基于自适应显著特征选择的动态加权平均行人识别模型[J]. 计算机工程与科学, 2017, 39(05): 935-943.
[14]	张娟,蒋芸,胡学伟,肖吉泽. 基于卷积受限玻尔兹曼机的医学图像分类新方法[J]. 计算机工程与科学, 2017, 39(02): 323-329.
[15]	马慧芳，周汝南，吉余岗，鲁小勇. 融合词语类别特征和语义的短文本分类方法[J]. 计算机工程与科学, 2017, 39(02): 399-404.