INTRODUCTION: Electrophysiological studies of human motor units can use various electromyographic techniques. Together with the development of new techniques for analysis and processing of bioelectric signals, motor unit action potential (MUAP) wavelet analysis represents an important change in the development of electromyographic techniques. METHODS: The proposed approach involves isolating single MUAPs, computing their scalograms, taking the maximum values of the scalograms in 5 selected scales, and averaging across MUAPs to give a single five-dimensional feature vector per muscle. After Support Vector Machine analysis, the feature vector is reduced to a single decision parameter that allows the subject to be assigned to 1 of 3 groups: myogenic, healthy, or neurogenic. The software is available as freeware. RESULTS: MUAP wavelet analysis yielded consistent results for the diagnostic index and muscle classification, with only 7 incorrect classifications out of a total of 1,015 samples. CONCLUSIONS: This proposed approach provides a sensitive and reliable method for evaluating and characterizing MUAPs.
INTRODUCTION: Electrophysiological studies of human motor units can use various electromyographic techniques. Together with the development of new techniques for analysis and processing of bioelectric signals, motor unit action potential (MUAP) wavelet analysis represents an important change in the development of electromyographic techniques. METHODS: The proposed approach involves isolating single MUAPs, computing their scalograms, taking the maximum values of the scalograms in 5 selected scales, and averaging across MUAPs to give a single five-dimensional feature vector per muscle. After Support Vector Machine analysis, the feature vector is reduced to a single decision parameter that allows the subject to be assigned to 1 of 3 groups: myogenic, healthy, or neurogenic. The software is available as freeware. RESULTS: MUAP wavelet analysis yielded consistent results for the diagnostic index and muscle classification, with only 7 incorrect classifications out of a total of 1,015 samples. CONCLUSIONS: This proposed approach provides a sensitive and reliable method for evaluating and characterizing MUAPs.