Computer system for identification and analysis of motor unit potential trains
Ł. Mazurkiewicz, Maria Piotrkiewicz · 2004
The paper presents a computerized system for identification and analysis of electromyographic signals recorded from muscle by needle electrodes. The system provides information on human motoneuron activity as time sequences of identified single motor unit potentials. This information can afterwards be statistically processed to obtain data on motoneuron properties in healthy subjects and in some neuromuscular diseases. A special attention has been paid to the description of the methods of spike identification, including extraction of spikes from the signal and automatic classification, which is based on the statistical pattern recognition approach. The system described is also an example of adaptation of modern computer technology for biomedical research.