Performance index assessment of intelligent computing methods in EMG-based neuromuscular diseases

Abstract

In the medical systems, there is a lack of determining and assessing the performance measure of the intelligent computing methods ICM deployed in the diagnosis of bioelectric signals EEG/EMG/ECG-based diseases. There have been few attempts for performance measure of mathematical models in medical computing. In this paper, we have developed a heuristic method for the assessment of performance measure in the diagnosis of EMG-based neuromuscular diseases. Firstly, we review the various ICM then we perform qualitative assessment of mathematical, algorithmic and heuristic content, data acquisition cost as well as medical consultancy cost of various parameter of EMG and non-EMG psychological, cognitive and muscular. The computational overhead CO of EMG parameters, overall computational overhead OCO and clinical consultancy cost CC are determined. Finally, performance index PI is computed based on the two overheads and clinical cost. A graph showing the comparative view of CO of EMG parameters, OCO and CC is plotted and PI is shown for all the methods.