Multi-class classification of biomechanical data: A functional LDA approach based on multi-class penalized functional PLS

被引:9
作者
Aguilera-Morillo, M. Carmen [1 ,2 ]
Aguilera, Ana M. [3 ]
机构
[1] Univ Carlos III Madrid, Escuela Politecn Super, Dept Stat, Avda Univ 30, Madrid 28911, Spain
[2] Univ Carlos III Madrid, Santander Big Data Inst UC3M BS, Madrid, Spain
[3] Univ Granada, Fac Ciencias, Dept Stat & OR & IEMath GR, Granada, Spain
关键词
functional data; Linear discriminant analysis; multi-class classification; PLS regression; P-spline penalty; PRINCIPAL-COMPONENT ANALYSIS; PARTIAL LEAST-SQUARES; LINEAR DISCRIMINANT-ANALYSIS; REGRESSION; GAIT; KINEMATICS; COORDINATION; ASYMMETRY; MODELS; PCA;
D O I
10.1177/1471082X19871157
中图分类号
O21 [概率论与数理统计]; C8 [统计学];
学科分类号
020208 ; 070103 ; 0714 ;
摘要
A functional linear discriminant analysis approach to classify a set of kinematic data (human movement curves of individuals performing different physical activities) is performed. Kinematic data, usually collected in linear acceleration or angular rotation format, can be identified with functions in a continuous domain (time, percentage of gait cycle, etc.). Since kinematic curves are measured in the same sample of individuals performing different activities, they are a clear example of functional data with repeated measures. On the other hand, the sample curves are observed with noise. Then, a roughness penalty might be necessary in order to provide a smooth estimation of the discriminant functions, which would make them more interpretable. Moreover, because of the infinite dimension of functional data, a reduction dimension technique should be considered. To solve these problems, we propose a multi-class approach for penalized functional partial least squares (FPLS) regression. Then linear discriminant analysis (LDA) will be performed on the estimated FPLS components. This methodology is motivated by two case studies. The first study considers the linear acceleration recorded every two seconds in 30 subjects, related to three different activities (walking, climbing stairs and down stairs). The second study works with the triaxial angular rotation, for each joint, in 51 children when they completed a cycle walking under three conditions (walking, carrying a backpack and pulling a trolley). A simulation study is also developed for comparing the performance of the proposed functional LDA with respect to the corresponding multivariate and non-penalized approaches.
引用
收藏
页码:592 / 616
页数:25
相关论文
共 61 条
[1]   Penalized versions of functional PLS regression [J].
Aguilera, A. M. ;
Aguilera-Morillo, M. C. ;
Preda, C. .
CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS, 2016, 154 :80-92
[2]   Comparative study of different B-spline approaches for functional data [J].
Aguilera, A. M. ;
Aguilera-Morillo, M. C. .
MATHEMATICAL AND COMPUTER MODELLING, 2013, 58 (7-8) :1568-1579
[3]   Using basis expansions for estimating functional PLS regression Applications with chemometric data [J].
Aguilera, Ana M. ;
Escabias, Manuel ;
Preda, Cristian ;
Saporta, Gilbert .
CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS, 2010, 104 (02) :289-305
[4]  
[Anonymous], 2013, Communications in Statistics. Simulation and Computation
[5]   Partial least squares for discrimination [J].
Barker, M ;
Rayens, W .
JOURNAL OF CHEMOMETRICS, 2003, 17 (03) :166-173
[6]   Functional linear discriminant analysis: a new longitudinal approach to the assessment of embryonic growth [J].
Bottomley, C. ;
Daemen, A. ;
Mukri, F. ;
Papageorghiou, A. T. ;
Kirk, E. ;
Pexsters, A. ;
De Moor, B. ;
Timmerman, D. ;
Bourne, T. .
HUMAN REPRODUCTION, 2009, 24 (02) :278-283
[7]   Effect of registration on cyclical kinematic data [J].
Crane, Elizabeth A. ;
Cassidy, Ruth B. ;
Rothman, Edward D. ;
Gerstner, Geoffrey E. .
JOURNAL OF BIOMECHANICS, 2010, 43 (12) :2444-2447
[8]   An anova test for functional data [J].
Cuevas, A ;
Febrero, M ;
Fraiman, R .
COMPUTATIONAL STATISTICS & DATA ANALYSIS, 2004, 47 (01) :111-122
[9]   PCA in studying coordination and variability: a tutorial [J].
Daffertshofer, A ;
Lamoth, CJC ;
Meijer, OG ;
Beek, PJ .
CLINICAL BIOMECHANICS, 2004, 19 (04) :415-428
[10]   Optimal Bayes classifiers for functional data and density ratios [J].
Dai, Xiongtao ;
Mueller, Hans-Georg ;
Yao, Fang .
BIOMETRIKA, 2017, 104 (03) :545-560