Some remarks on computer-aided clinical electromyography based on experience with the Polish ANOPS minicomputer.

The author evaluates the application of the Polish minicomputer Analyser of Noise of Periodic Signals (ANOPS) in practical electromyography (EMG) and compares the diagnostic yield of ANOPS with that of manual quantitative EMG. The advantages of ANOPS--at least in the past--include speed, large samples, new data, reduced examiner bias, quantified evaluation of maximal effort record and online recording. The most important limitations of ANOPS are absence of complex potentials due to distortion of signals, loss of phase evaluation, and lack of comparability of amplitude with other methods. ANOPS is compared with other automated methods applied to evaluation of interference pattern and parameters of single motor unit potentials (MUPs). Most of the latter are based on pattern recognition. ANOPS does not meet all of the criteria which an automatic method should meet to be accepted in EMG practice. The most important problem is the distortion of properties of the EMG signal. Further search for computer-aided EMG methods and their careful testing in practice are necessary.