응용통계연구 (The Korean Journal of Applied Statistics)

  • 제34권2호
  • /
  • Pages.127-139
  • /
  • 2021
  • /
  • 1225-066X(pISSN)
  • /
  • 2383-5818(eISSN)
한국통계학회 (The Korean Statistical Society)
권호 표지
DOI QR코드

DOI QR Code

로버스트 그룹 독립성분분석

Robust group independent component analysis

  • Kim, Hyunsung (Department of Statistics, Chung-Ang University) ;
  • Li, XiongZhu (Department of Statistics, Chung-Ang University) ;
  • Lim, Yaeji (Department of Statistics, Chung-Ang University)
  • 투고 : 2020.11.03
  • 심사 : 2021.01.04
  • 발행 : 2021.04.30
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

독립성분분석은 혼합 데이터로부터 독립된 신호들을 분리해내는 대표적인 통계적 방법론이며, 그룹 독립성분분석은 독립성분분석을 여러 개체에 적용할 수 있도록 확장한 방법론이다. 그룹 독립성분분석은 기능적 자기 공명 영상 데이터에 활용되어 의학적으로 유의미한 결과를 줌이 알려져있다. 그러나 자기 공명 영상 스캔에서 흔히 일어나는 이상치가 포함되어 있는 경우, 기존의 그룹 독립성분분석은 그 효과가 떨어짐이 알려져있다. 본 연구에서는 ROBPCA 기반의 로버스트한 그룹 독립성분분석 방법론을 제안하였다. 시뮬레이션과 실제 자료 분석을 통해 제안한 방법과 기존 방법을 비교하였고, 그 결과 제안한 방법론의 로버스트성을 입증했다.

Independent Component Analysis is a popular statistical method to separate independent signals from the mixed data, and Group Independent Component Analysis is an its multi-subject extension of Independent Component Analysis. It has been applied Functional Magnetic Resonance Imaging data and provides promising results. However, classical Group Independent Component Analysis works poorly when outliers exist on data which is frequently occurred in Magnetic Resonance Imaging scanning. In this study, we propose a robust version of the Group Independent Component Analysis based on ROBPCA. Through the numerical studies, we compare proposed method to the conventional method, and verify the robustness of the proposed method.

키워드

과제정보

이 논문은 2019년도 중앙대학교 CAU GRS 지원에 의하여 작성되었음.

참고문헌

  1. Bai P, Shen H, Huang X, and Truong Y (2008). A supervised singular value decomposition for independent component analysis for fMRI, Statistica Sinica, 18, 1233-1252.
  2. Biswal BB and Ulmer JL (1999). Blind source separation of multiple signal sources of fMRI data sets using independent component analysis, Journal of Computer Assisted Tomography, 23, 265-271. https://doi.org/10.1097/00004728-199903000-00016
  3. Bordier C, Dojat M, and de Micheaux PL (2011). Temporal and spatial independent component analysis for fMRI data sets embedded in the AnalyzeFMRI r package, Journal of Statistical Software, 44, 1-24.
  4. Bulut H and Oner Y (2017). The evaluation of socio-economic development of development agency regions in Turkey using classical and robust principal component analyses, Journal of Applied Statistics, 44, 2936- 2948. https://doi.org/10.1080/02664763.2016.1267115
  5. Calhoun VD, Adali T, Pearlson GD, and Pekar JJ (2001). A method for making group inferences from functional MRI data using independent component analysis, Human Brain Mapping, 14, 140-151. https://doi.org/10.1002/hbm.1048
  6. Comon P (1994). Independent component analysis, a new concept?, Signal Processing, 36, 287-314. https://doi.org/10.1016/0165-1684(94)90029-9
  7. De Klerk J (2015). Time series outlier detection using the trajectory matrix in singular spectrum analysis with outlier maps and ROBPCA, South African Statistical Journal, 49, 61-76.
  8. Hubert M, Rousseeuw PJ, and Vanden Branden K (2005). ROBPCA:A new approach to robust principal component analysis, American Statistical Association and the American Society for Quality, 47.
  9. Hyvarinen A (1999). Fast and robust fixed-point algorithms for independent component analysis, IEEE Transactions on Neural Networks, 10, 626-634. https://doi.org/10.1109/72.761722
  10. Hyvarinen A and Oja E (2000). Independent component analysis: algorithms and applications, Journal of Neural Networks, 13, 411-430. https://doi.org/10.1016/S0893-6080(00)00026-5
  11. Hyvarinen A (2013). Independent component analysis: recent advances, Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 371, 20110534. https://doi.org/10.1098/rsta.2011.0534
  12. Langlois D, Chartier S, and Gosselin D (2010). An introduction to independent component analysis: infoMax and fastICA algorithms, Tutorials in Quantitative Methods for Psychology, 6, 31-38. https://doi.org/10.20982/tqmp.06.1.p031
  13. McKeown MJ and Sejnowski TJ (1998). Independent component analysis of FMRI data: examining the assumptions, Human Brain Mapping 6, 368-372. https://doi.org/10.1002/(SICI)1097-0193(1998)6:5/6<368::AID-HBM7>3.0.CO;2-E
  14. McKeown MJ, Jung TP, Makeig S, Brown G, Kindermann SS, Lee TW, and Sejnowski TJ (1998). Spatially independent activity patterns in functional MRI data during the stroop color-naming task. In Proceedings of the National Academy of Sciences, 95, 803-810. https://doi.org/10.1073/pnas.95.3.803
  15. Meinecke FC, Harmeling S, and Muller KR (2004). Robust ICA for super-Gaussian sources, In International Conference on Independent Component Analysis and Signal Separation, 17-224, Springer, Berlin, Heidelberg.
  16. Pruim RH, Mennes M, van Rooij D, Llera A, Buitelaar JK, and Beckmann CF (2015). ICA-AROMA: A robust ICA-based strategy for removing motion artifacts from fMRI data, Neuroimage, 112, 267-277. https://doi.org/10.1016/j.neuroimage.2015.02.064
  17. Rachakonda S, Egolf E, Correa N, and Calhoun V (2007). Group ICA of fMRI toolbox (GIFT) manual. Available from: http://www.nitrc.org/docman/view.php/55/295.
  18. Rousseeuw PJ (1984). Least median of squares regression, Journal of the American Statistical Association, 89, 871-880. https://doi.org/10.1080/01621459.1984.10477105
  19. Rousseeuw PJ and Driessen KV (1999). A fast algorithm for the minimum covariance determinant estimator, Technometrics, 41, 212-223. https://doi.org/10.1080/00401706.1999.10485670
  20. Todorov V (2009). Rrcov: scalable robust estimators with high breakdown point, r package version 0.5-03. Available from: http://CRAN.R-project.org/package=rrcov.