构建基于小波熵的自训练半监督支持向量机分类模型评价老年人步态

doi:10.3969/j.issn.0258-8021. 2013. 05.011

摘要
图/表
参考文献
相关文章 (15)

全文: PDF (571 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要研究应用半监督学习算法分析未标注步态数据评价老年人步态，提出基于小波熵的自训练半监督支持向量机步态分类模型，通过小波熵从未标注步态数据中选取为每次自训练步态分类模型所需最具信息量的标注样本，有效获取步态数据类别间和步态数据内在的“有价值”的步态变异信息，提高步态分类器的泛化性能。首先采用10名老年人和10名青年人步态数据构建支持向量机分类模型，然后对120名不同年龄组未标注步态数据分类预测，依据小波熵选取样本数据，逐步添加更新步态样本训练集，自训练支持向量机分类模型。实验结果表明，本算法较准确鉴别青年和老年人步态模式（分类正确率90%），比基于有监督学习的支持向量机步态分类算法正确率提高近5%，有效改善支持向量机步态分类算法性能，有望为临床提供一个评价老年人步态的新方法。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	吴建宁^* 伍滨

关键词 ：步态分析, 半监督学习, 支持向量机, 小波熵, 老年人

Abstract：This work investigated the application of the semisupervised learning algorithm to analysis of the unlabeled gait data for evaluating the elderly gait. The novel gait classification algorithm that the selftraining semisupervised support vector machine(SVM) based on wavelet entropy for the discrimination between the young and elderly gait pattern was addressed. In the selftraining, wavelet entropy was employed to obtain labeled samples from unlabeled dataset. The most valuable information related to the gait change was acquired for current gait classifier model, which obviously improved the gait classification performance of SVM. The labeled gait sample datasets including 10 young and 10 elderly participants were used to develop SVM that was employed to classify the unlabelled gait dataset from 120 subjects of different age groups. The new labeled gait data, obtained by our defined wavelet entropy, were selected and constructed the new sample train dataset for developing selftraining SVM. The experimental results showed that the accuracy of our proposed algorithm is 90% in recognization of the young and elderly gait pattern. Furthermore, the accuracy of our proposed algorithm was increased approximately 5% compared with that of the classification algorithm by the supervised support vector machine, suggesting that our proposed technique can obtain more information related to gait change from the labeled and unlabeled gait dataset, and provides a new tool for assessment of elder gait.

Key words： gait analysis semisupervised learning support vector machine wavelet entropy elderly people

基金资助:福建省自然科学基金（2013J01220）；福建省教育厅(B类)项目（JB12032）

引用本文:

吴建宁^* 伍滨. 构建基于小波熵的自训练半监督支持向量机分类模型评价老年人步态[J]. 中国生物医学工程学报, 2013, 32(5): 588-594.
WU Jian Ning^* WU Bin. The SelfTraining SemiSupervised Support Vector Machine Based on Wavelet Entropy for the Evaluation of the Elderly Gait. journal1, 2013, 32(5): 588-594.

链接本文:

http://cjbme.csbme.org/CN/10.3969/j.issn.0258-8021. 2013. 05.011 或 http://cjbme.csbme.org/CN/Y2013/V32/I5/588

［1］中华人民共和国卫生部疾病预防控制局. 老年人跌倒干预技术指南［EB/OL］. http://www.moh.gov.cn, 2011-09-06/2012-11-20.
［2］Giannini S, Catani F, Benedetti MG, et al. Gait analysis: Methodologies and clinical applications ［M］ // Benedetti MG, eds. Terminology, Parameterization and Normalization in Gait Analysis. Amsterdam: IOS Press, 1994:65-88.
［3］Brach JS, Perera S, Studenski S, et al. Meaningful change in measures of gait variability in older adults ［J］. Gait and Posture, 2010, 31(2): 175-179.
［4］Nigg BM, Skleryk BN. Gait characteristics of the elderly ［J］. Clinical Biomechanics, 1988, 3(2):79-87.
［5］Franois P, Hélène C, Réjean H. Gait in the elderly ［J］. Gait and Posture, 1997, 5(2):128-135.
［6］Chau T. A review of analytical techniques for gait data. Part 2: Neural network and wavelet methods ［J］. Gait and Posture, 2001, 13: 102-120.
［7］Wu Jianning, Wang Jue, Liu Li. Feature extraction via KPCA for classification of gait patterns ［J］. Human Movement Science, 2007, 26(3):393-411.
［8］Begg RK, Palaniswami M, Owen B. Support vector machines for automated gait classification ［J］. IEEE Trans on Biomedical Engineering, 2005, 52(9): 828-838.
［9］Khandoker AH, Lai DT, Begg RK, et al. Waveletbased feature extraction for support vector machines for screening balance impairments in the elderly ［J］. IEEE Trans on Neural Systems and Rehabilitation Engineering, 2007, 15(4):587-97.
［10］Begg RK, Kamruzzaman J. A machine learning approach for automated recognition of movement patterns using basic, kinetic and kinematic gait data ［J］. Journal of Biomechanics, 2005, 38: 401-408.
［11］Reginaldo K, Fukuchi B, Bjoern M, et al. Support vector machines for detecting agerelated changes in running kinematics ［J］. Journal of Biomechanics, 2011, 44(3):540-542.
［12］尧德中, 刘铁军, 雷旭, 等. 基于脑电的脑-机接口：关键技术和应用前景［J］. 电子科技大学学报, 2009, 38(5): 550-553.
［13］王金甲，贾敏. 标签均值的自训练半监督支持向量机用于EEG分类［J］. 中国生物医学工程学报, 2011, 30(5):666-672.
［14］Chapelle O, Scholkopf B, Zien A. Semisupervised Learning ［M］ // Joachims T,eds. Transductive Support Vector Machines. Cambridge: MIT Press, 2006:101-112.
［15］吴建宁, 伍滨, 王珏. 用小波熵评价人体步态变化［J］. 航天医学与医学工程, 2009, 22（6）:437-442.
［16］Begg RK, Rahman SM. A method for the reconstruction of ground reaction forcetime characteristics during gait from force platform recordings of simultaneous foot falls ［J］. IEEE Trans on Biomedical Engineering, 2000, 47:547-551.
［17］Chang CC, Lin CJ. LIBSVM: a library for support vector machine ［EB/CP］. http://www.csie.ntu.edu.tw/～cjlin/libsvm, 2009-05-01/2012-11-20.