ART2 Neural Network for Surface EMG Decomposition

Extraction of individual Motor Unit Action Potentials (MUAPs) from a surface ElectroMyoGram (EMG) is an essential but challenging task for clinical study and physiological investigation. This paper presents an automatic decomposition of surface EMGs using a self-organised ART2 neural network. In our approach, MUAP peaks are first detected using a Weighted Low-Pass Differential (WLPD) filter. A modified ART2 network is then utilised to classify MUAPs based on MUAP waveforms and firing time information. Individual MUAP trains are identified from real surface EMG signals recorded during weak contraction, and also from simulated surface EMGs. The firing statistics and the waveforms of individual MUAPs are then extracted. A number of computer tests on 50 simulated and real surface EMGs of limb muscles show that up to five MUAP trains can be effectively extracted, with their waveforms and firing parameters estimated. Being able to decompose real surface EMGs has essentially demonstrated the potential applications of our approach to the non-invasive diagnosis of neuromuscular disorders.

[1]  Shaojun Xiao,et al.  Estimation of motor unit firing statistics from surface EMG , 1998, Proceedings of the 20th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. Vol.20 Biomedical Engineering Towards the Year 2000 and Beyond (Cat. No.98CH36286).

[2]  M.H. Hassoun,et al.  NNERVE: Neural Network Extraction of Repetitive Vectors for Electromyography. I. Algorithm , 1994, IEEE Transactions on Biomedical Engineering.

[3]  P. Rosenfalck.  Intra- and extracellular potential fields of active nerve and muscle fibres. A physico-mathematical analysis of different models. , 1969, Acta physiologica Scandinavica. Supplementum.

[4]  George S. Moschytz,et al.  A New Framework and Computer Program for Quantitative EMG Signal Analysis , 1984, IEEE Transactions on Biomedical Engineering.

[5]  Carlo J. De Luca,et al.  Physiology and Mathematics of Myoelectric Signals , 1979 .

[6]  Shiro Usui,et al.  Digital Low-Pass Differentiation for Biological Signal Processing , 1982, IEEE Transactions on Biomedical Engineering.

[7]  H. Broman.  Automatic decomposition of myoelectric signals , 1988, Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[8]  Isak Gath,et al.  Techniques for Improving the Selectivity of Electromyographic Recordings , 1976, IEEE Transactions on Biomedical Engineering.

[9]  Steen Andreassen,et al.  Recording from a Single Motor Unit During Strong Effort , 1978, IEEE Transactions on Biomedical Engineering.

[10]  S. Xiao.  Propagating point-source response of active fibers , 1995, Proceedings of 17th International Conference of the Engineering in Medicine and Biology Society.

[11]  Kevin C. McGill,et al.  Automatic decomposition electromyography , 1985 .

[12]  Kevin C. McGill,et al.  Automatic Decomposition of the Clinical Electromyogram , 1985, IEEE Transactions on Biomedical Engineering.

[13]  M.H. Hassoun,et al.  NNERVE: Neural Network Extraction of Repetitive Vectors for Electromyography. II. Performance analysis , 1994, IEEE Transactions on Biomedical Engineering.

[14]  Ronald S. Lefever,et al.  A Procedure for Decomposing the Myoelectric Signal Into Its Constituent Action Potentials - Part I: Technique, Theory, and Implementation , 1982, IEEE Transactions on Biomedical Engineering.

[15]  Charles Leave.  Neural Networks: Algorithms, Applications and Programming Techniques , 1992 .

[16]  D. Stashuk,et al.  Automatic decomposition of selective needle-detected myoelectric signals , 1988, IEEE Transactions on Biomedical Engineering.