Application of Multi-Dimensional Principal Component Analysis to Medical Data

Authors: Naoki Yamamoto, Jun Murakami, Chiharu Okuma, Yutaro Shigeto, Satoko Saito, Takashi Izumi, Nozomi Hayashida

Abstract:

Multi-dimensional principal component analysis (PCA) is the extension of the PCA, which is used widely as the dimensionality reduction technique in multivariate data analysis, to handle multi-dimensional data. To calculate the PCA the singular value decomposition (SVD) is commonly employed by the reason of its numerical stability. The multi-dimensional PCA can be calculated by using the higher-order SVD (HOSVD), which is proposed by Lathauwer et al., similarly with the case of ordinary PCA. In this paper, we apply the multi-dimensional PCA to the multi-dimensional medical data including the functional independence measure (FIM) score, and describe the results of experimental analysis.

Keywords: multi-dimensional principal component analysis, higher-order SVD (HOSVD), functional independence measure (FIM), medical data, tensor decomposition

Digital Object Identifier (DOI): doi.org/10.5281/zenodo.1061360

References:

[1] Lindsay I. Smith, "A tutorial on principal components analysis," www.cs. otago.ac.nz/osc453/student_tutorials/principal_components.pdf, 2002.
[2] Hervé Abdi and Lynne J. Williams, "Principal components analysis," Wiley Interdisciplinary Reviews: Computational Statistics, vol.2, pp. 433-450, 2010.
[3] Naoki Yamamoto, Jun Murakami, and Chiharu Okuma, "Application of matrix PCA to school record analysis," Proceedings of the IEICE General Conference (in Japanese), D-15-14, 2011.
[4] Yana Mulyana, Naoki Yamamoto, Jun Murakami, and Chiharu Okuma, "Analysis of multi-dimensional data by using multi-dimensional principal component analysis ´╝ìapplication to Monthly Labour Survey data," Proceedings of the 25th Technology Exchange Meeting between Industry, Academia, and Government in Kumamoto (in Japanese), pp.68-69, 2011.
[5] Kohei Inoue, Kenji Hara, and Kiichi Urahama, "Matrix principal component analysis for image compression and recognition," Proceedings of the 1st Joint Workshop on Machine Perception and Robotics (MPR), pp115-120, 2005.
[6] Jieping Ye, Revi Janardan, and Qi Li, "GPCA: an efficient dimension reduction scheme for image compression and retrieval," Proceedings of the 10th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp.354-363, 2004.
[7] Rui Xu and Yen-Wei Chen, "Generalized N-dimensional principal component analysis (GND-PCA) and its application on construction of statistical appearance models for medical volumes with fewer samples," Neurocomputing, vol.72, no.10-12, pp.2276-2287, 2009.
[8] Kenneth J. Ottenbacher, Yungwen Hsu, Carl V. Granger, and Roger C. Fiedler, "The reliability of the functional independence measure: A quantitative review," Archives of Physical Medicine and Rehabilitation, vol.77, pp.1226-1232, 1996.
[9] Lisbeth Claesson and Elisabeth Svensson, "Measures of order consistency between paired ordinal data: application to the Functional Independence Measure and Sunnaas index of ADL," Journal of Rehabilitation Medicine, vol. 33, 2001.
[10] James W. Davis and Hui Gao, "An expressive three-mode principal components model for gender recognition," Journal of Vision, vol.4, pp. 362-377, 2004.
[11] Lieven De Lathauwer, Bart De Moor, and Joos Vandewalle, "A multilinear singular value decomposition," SIAM Journal on Matrix Analysis and Applications, vol.21, no.4, pp.1253-1278, 2000.
[12] Gene H. Golub and Christian H. Reinsch, "Singular value decomposition and least squares solution," Numerical Mathematics, vol. 14, pp.403-420, 1970.
[13] George E. Forsythe, Michael A. Malcolm, and Cleve B. Moler, Computer Methods for Mathematical Computations, chap.9, Englewood Cliffs, Prentice-Hall, New Jersey, 1977.
[14] Chiharu Okuma, Jun Murakami, and Naoki Yamamoto, "Comparison between higher-order SVD and third-order orthogonal tensor product expansion," International Journal Electronics, Communications and Computer Engineering, vol.1, no.2, pp.131-137, 2009.
[15] Chiharu Okuma, Naoki Yamamoto, and Jun Murakami, "An improved algorithm for calculation of the third-order orthogonal tensor product expansion by using singular value decomposition," International Journal Electronics, Communications and Computer Engineering, vol.2, no.1, pp.11-20, 2010.
[16] Tobias Heimann, Ivo Wolf, Tomos G. Williams, and Hans P. Meinzer, "3D active shape models using gradient descent optimization of description length," in Proceedings of the IPMI, pp.566-567, Springer, 2005.
[17] Michael E. Wall, Andreas Rechtsteiner, and Luis M. Rocha, "Singular value decomposition and principal component analysis, " in A Practical Approach to Microarray Data Analysis (D. P. Berrar, W. Dubitzky, M. Granzow eds.), Kluwer: Norwell, Massachusetts, pp.91-109, 2003.