::Conclusion
This method demonstrates the utility of a Bayesian approach to analysis of EMG data, in its ability to make statistically valid estimates of the probability of muscle activity/inactivity, and thus specific muscle activation patterns. Probablistic estimates of temporal activation patterns (on and off-set times) not only allow statistical comparisons of relevant quantities, but also reduce the amount of unknown error associated with such measurements.
We further show the ability of modern Markov chain Monte Carlo simulation to estimate posterior distributions of parameters in complex models of EMG data.
The application demonstrated here could be used to address many investigational goals. These quantitative measurements could serve as meaningful clinical measures of specific neuropathologies, normal locomotor muscle activation patterns, and gait rehabilitative progress and strategies.
::References
Green, P. J. (1995). Reversible jump Markov chain Monte Carlo computation and Bayesian model determination. Biometrika 82, 711-732.
Hastings, W. K. (1970). Monte Carlo sampling methods using Markov chains and their applications. Biometrika 57, 97-109.
Johnson, T. D., Elashoff, R. M., and Harkema, S. J. (submitted). Bayesian change-point analysis of muscular coordination from electromyographic data. Biometrics.
URL: http://www.biomath.medsch.ucla.edu/faculty/tjohnson/