Abstract
A certainty-based classification algorithm is described, which comprises part of a clinically used EMG signal decomposition system. This algorithm classifies a candidate motor unit action potential (MUAP) to the motor unit action potential train (MUAPT) that produces the greatest estimated certainty, provided this maximal certainty is above a given threshold. The algorithm is iterative, such that the certainty with which assignments are made increases with each pass through the data, and it has specific stopping criteria. The performance and sensitivity (to the assignment threshold) of the Certainty algorithm and an iterative minimum Euclidean distance (MED) algorithm are compared by classifying sets of MUAPs detected in real concentric needle-detected EMG signals, using a range of assignment thresholds for each algorithm. With regard to MUAP assignment and error rates, the Certainty algorithm consistently provides better mean results and, more importantly, less variable results than the MED algorithm. The Certainty algorithm can provide mean assignment and error rates of 80.8 and 1.5%, respectively, with a maximum error rate of 3.2%; the MED algorithm can provide mean assignment and error rates of 80.3 and 3.3%, respectively, with a maximum error rate of 6.5%. The Certainty algorithm is relatively insensitive to the certainty threshold used, can consistently differentiate between similarly shaped MUAPs from different MUAPTs, and can make correct classifications despite biological shape variability, background noise and signal shape non-stationarity.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Basmajian, J. V., andDeLuca, C. J. (1985): ‘Muscles alive (The Williams & Wilkins Co., Baltimore, USA)
Barkhaus, P. E., andNandedkar, S. D. (1993): ‘Clinical evaluation of a decomposition algorithm for motor unit action potential analysis’,Muscle Nerve,16, p. 1085
Bhatnagar, R. K., andKanal, L. N. (1985): ‘Handling uncertain information: a review of numeric and non-numerica methods’inL. N.Kanal, andLemmerJ. F. (Eds.) ‘Uncertainty in artificial intelligence’, (North-Holland, New York) pp. 3–26
Bischoff, C., Stalberg, E. V., Falck, B., andEeg-Olofsson, K. E. (1994): ‘Reference values of motor unit action potentials obtained with multi-MUAP analysis’,Muscle Nerve,17, pp. 842–851
Broman, H. (1988): ‘Knowledge-based signal processing in the decomposition of myoelectric signals’,IEEE Eng. Med. Biol. Mag., pp. 24–28
Bunn, D. W. (1984): ‘Applied decision analysis’ (McGraw-Hill, New York)
Conwit, R. A., Tracy, B., Jamison, C., McMugh, M., Stashuk, D. W., Brown, W. F., andMetter, E. J. (1997): ‘Decomposition-enhanced spike triggered averaging: contraction level effects’,Muscle Nerve,20, pp. 976–982
Doherty, T., Simmons, Z., O'Connell, B., Felice, K., Conwit, R., Ming Chan, K., Komori, T. Brown, T., Stashuk, D., andBrown, W. (1995): ‘Methods for estimating the numbers of motor units in human muscles’,J. Clin. Neurophysiol.,12, pp. 565–584
Etawil, H. A. Y., andStashuk, D. W. (1996): ‘Resolving super-imposed motor unit action potentials’,Med. Biol. Eng. Comput.,34, (1), pp. 33–40
Etawil, H. A. Y. (1994): ‘Motor unit action potentials: discovering temporal relations of their trains and resolving their superpositions’. Master's Thesis, University of Waterloo
Gerber, A. S., andFigueredo, R. J. (1983): ‘MAP detection and ML estimation of signals overlapping in time and frequency for both Gaussian and Laplacian noise. Proc. IEEE ICASSP Conf., Boston, USA, pp. 603–606
Gerber, A. S., Studer, R. M., Figueredo, R. J., andMoshytz, G. S. (1984): ‘A new framework and computer program for quantitative EMG sigmal analysis’,IEEE Trans.,BME-31, pp. 857–863
Golub, G. H., andVan Loan C. F. (1983): ‘Matrix computations’ (The Johns Hopkins University Press, Baltimore, Maryland, USA)
Guiheneuc, P., Calanel, J., Doncarli, C., Gitton, D., andMichel, C. (1983): ‘Automatic detection and pattern recognition of single motor unit potentials in needle EMG’, inDesmedt, J. E. (Ed.): ‘Computer-aided electromyography’ (Progr. Clin. Neurophysiol., vol. 10) (Karger, Basel, Switzerland) pp. 73–127
Haas, W. F., andMeyer, M. (1989): ‘An automated EMG decomposition system for routine clinical examination and clinical research, ARTMUP—automatic recognition and tracking of motor unit potentials’in:Desmedt, J. E., (Ed.): ‘Computeraided electromyography and expert systems’ (Clin. Neurophysiol. Updates, vol. 2) (Elsevier) pp. 67–81
Hassoun, M. H., Wang, C., andSpitzer, A. R. (1994): ‘NNERVE: neural network extraction of repetitive vectors for electromyography—Part I: algorithm’,IEEE Trans.,BME-41, (11), pp. 1039–1052
Howard, J. E., McGill, K. C., andDorfman, L. J. (1988): ‘Properties of motor unit action potentials recorded with concentric and monopolar needle electrodes: ADEMG analysis’,Muscle Nerve,11, pp. 1051–1055
LeFever, R. S., andDeLuca, C. J. (1982): ‘Procedure for decomposing the myoelectric signal into its constituent action potentials—Part I: technique, theory and implementation’,IEEE Trans.,BME-29, (3), pp. 149–157
Loudon, G. H., Jones, N. B., andSehmi, A. S. (1992): ‘New signal and processing techniques for the decomposition of EMG signals’,Med. Biol. Eng. Comput.,30, (6), pp. 591–599
Mambrito, B., andDeLuca, C. J. (1983): ‘Acquisition and decomposition of the EMG signal’inDesmedt, J. E. (Ed.): ‘Computeraided electromoygraphy’ (Progr. in Clin. Neurophysiol.), Vol. 10, (Karger, Basel, Switzerland) pp. 52–72
McGill, K. (1984): ‘A method for quantitating the clinical electromyogram’ PhD Thesis, Stanford University, Stanford, California
McGill, K., Cummins, K. L., andDorfman, L. J. (1985): ‘Automatic decomposition of the clinical electromyogram’,IEEE Trans.,BME-32, (7), pp. 470–477
Nandedkar, S. D., andBarkhaus, P. E. (1993): ‘Motor unit action potential anlaysis using a decomposition algorithm’,Muscle Nerve,16, p. 1086
Paoli, G. M. (1993): ‘Estimating certainty in classification of motor unit action potentials’. Master's Thesis, University of Waterloo
Paoli, G. M., andStashuk, D. W. (1993): ‘Supervised classification of motor unit action potentials based on estimates of classification certainty’. Proc. 15th Ann. IEEE Engineering in Medicine and Biology Society Conf., San Diego, California, pp. 1211–1212
Stalberg, E. V., Bischoff, C., andFalck, B. (1994): ‘Outliers a way to detect abnormality in quantitative EMG’,Muscle Nerve,17, pp. 392–399
Stalberg, E. V., andSonoo, M. (1994): ‘Assessment of variability in the shape of the motor unit action potential, the “jiggle” at consecutive discharges’,Muscle Nerve,17, pp. 1135–1144
Stashuk, D. W., andde Bruin, H. (1988): ‘Automatic decomposition of selective needle-detected myoelectric signals’,IEEE Trans.,BME-35, (1), pp. 1–10
Stashuk, D. W., andNaphan, R. K. (1992): ‘Probabilistic inference-based classification applied to myoelectric signal decomposition’,IEEE Trans.,BME-39, (4), pp. 346–355
Stashuk, D. W., andBrown, W. F. (1994): ‘Decomposition enhanced spike-triggered averaging: an improved method for estimating motor unit numbers in proximal muscles’,Muscle Nerve,17, p. 1098
Stashuk, D. W., andQu, Y. (1996a): ‘Adaptive motor unit action potential clustering using shape and temporal information’,Med. Biol. Eng. Comput.,34 (1), pp. 41–49
Stashuk, D. W., andQu, Y. (1996b): ‘Robust method for estimating motor unit firing-pattern statistics’,Med. Biol. Eng. Comput. m34, (1), pp. 50–57
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Stashuk, D., Paoli, G.M. Robust supervised classification of motor unit action potentials. Med. Biol. Eng. Comput. 36, 75–82 (1998). https://doi.org/10.1007/BF02522861
Received:
Accepted:
Issue Date:
DOI: https://doi.org/10.1007/BF02522861