Model Based Decomposition of MUAPS Into Their Constituent SFEAPS

  • Maria Georgiou Xyda

    Student thesis: Master's Thesis

    Abstract

    The motor unit action potential (MUAP) represents the spatial and temporal summation of single fibre extracellular action potentials (SFEAPS) generated from the same motor unit. MUAP morphology is affected by structural reorganisation of the motor unit that takes place due to disorders affecting peripheral nerve and muscle. MUAP features extracted in the time domain like duration, amplitude and phases are extensively used by the neurophysiologist for the assessment of neuromuscular disorders. In this study, a model based decomposition of MUAPS into their constituent SFEAPS is investigated. The aim of this study has been to develop a system that will give to the neurophysiologist a visualization of an "estimated" structural organisation of the motor unit, which includes information about number of fibres, fibre distribution and positioning and fibre diameter. This consists of the inverse problem in EMG.

    The mathematical model developed in two dimensions by Dimitrova and Dimitrov (Dimitrova 1974; Dimitrov 1987) was used to generate SFEAPS. In addition this model was extended to three dimensions. Two-dimensional and three-dimensional models of normal MUAPS consisting of 1, 4, 5, 10 and 50 fibres respectively were developed and decomposed for a small recoding radius up to 5 mm. Also three-dimensional models were developed and decomposed for myopathy in cases of variability in diameter, loss of fibres and reinnervation. The non linear least squares optimization procedure based on the Levenberg-Marquardt algorithm was used to obtain a solution to the MUAP decomposition problem i.e. fibre distribution, positioning and diameter. Using this method, a satisfactory solution to the decomposition problem was obtained. Future work will investigate the usefulness of the proposed analysis on MUAPS recorded from normal subjects and subjects suffering with neuromuscular disorders.
    Date of AwardNov 2006
    Original languageEnglish

    Cite this

    '