Wavelet-Based Multifractal Analysis of Human Balance

AbstractCenter of pressure (COP) traces have been used to investigate the dynamics of human balance. In this paper we employ a wavelet-based multifractal methodology to identify structural differences in mediolateral and anterioposterior sway between COP traces of healthy and Parkinson's patients. Two statistical techniques are used to summarize the differences in multifractal spectrum (MFS) for both groups. The first technique is a multivariate repeated measures analysis on estimated MF spectra for subjects. The second technique obtains two characteristic measures from each subject's estimated MFS: (i) location and (ii) half-width of the spectrum. These measures present an intuitive summary of the MFS for each subject, allowing for statistical comparisons between the two groups. Both analyzes lead to significant discrimination between Parkinson versus healthy subject's MFS. We find that COP time series of Parkinson patients exhibit a greater degree of roughness as compared to healthy subjects' COP traces. Furthermore, MFS for Parkinson patients are narrower, suggesting a reduction in complexity as compared to the healthy group. The methodology presented here may be helpful in development of clinically relevant measures, including the assessment of severity of conditions as the measures developed here correlate with standard severity measures. © 2002 Biomedical Engineering Society. PAC2002: 8719Bb, 8719St, 0250Sk, 8719Xx, 0545Df, 0230Uu

[1]  Thomas Stretch Dowse,et al.  MUSCLE AND NERVE , 1903 .

[2]  I. Daubechies.  Ten Lectures on Wavelets , 1992 .

[3]  E. Bacry,et al.  Singularity spectrum of fractal signals from wavelet analysis: Exact results , 1993 .

[4]  J. Collins,et al.  Random walking during quiet standing. , 1994, Physical review letters.

[5]  E. Weibel,et al.  Fractals in Biology and Medicine , 1994 .

[6]  Emmanuel Bacry,et al.  THE THERMODYNAMICS OF FRACTALS REVISITED WITH WAVELETS , 1995 .

[7]  R. Mantegna,et al.  Fractals in biology and medicine. , 1995, Chaos, solitons, and fractals.

[8]  H. Keselman,et al.  Analysing multivariate repeated measures designs when covariance matrices are heterogeneous , 1997 .

[9]  R. Baraniuk,et al.  A simple statistical analysis of wavelet-based multifractal spectrum estimation , 1998, Conference Record of Thirty-Second Asilomar Conference on Signals, Systems and Computers (Cat. No.98CH36284).

[10]  Rudolf H. Riedi,et al.  This is Printer Multifractal Processes , 1999 .

[11]  Patrice Abry,et al.  A Wavelet-Based Joint Estimator of the Parameters of Long-Range Dependence , 1999, IEEE Trans. Inf. Theory.

[12]  Benoit B. Mandelbrot,et al.  Multifractals and 1/f noise : wild self-affinity in physics (1963-1976) : selecta volume N , 1999 .

[13]  B. Mandelbrot.  Multifractals And 1/F Noise , 1999 .

[14]  Carson C. Chow,et al.  Assessing muscle stiffness from quiet stance in Parkinson's disease , 1999, Muscle & nerve.

[15]  U. Ruttimann,et al.  Determining the Hurst exponent of fractal time series and its application to electrocardiographic analysis , 1999, Comput. Biol. Medicine.

[16]  Vladimir M. Zatsiorsky,et al.  On the fractal properties of natural human standing , 2000, Neuroscience Letters.

[17]  S Thurner,et al.  Scaling-violation phenomena and fractality in the human posture control systems. , 2000, Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics.

[18]  J W Błaszczyk,et al.  Postural stability and fractal dynamics. , 2001, Acta neurobiologiae experimentalis.