A Signal Processing Approach to Practical Neurophysiology : A Search for Improved Methods in Clinical Routine and Research

摘要: Signal processing within the neurophysiological field is challenging and requires short time reliable results. In this thesis, three main problems are considered.First, a modified line source model for simulation of muscle action potentials (APs) presented. It formulated in continuous-time as convolution muscle-fiber dependent transmembrane current an electrode weighting (impedance) function. discretization model, Nyquist criterion addressed. By applying anti-aliasing filtering, it possible to decrease frequency while retaining accuracy. Finite length fibers incorporated through simple transformation The presented suitable modeling large motor units.Second, possibility discerning individual AP components concentric needle electromyogram (EMG) explored. Simulated unit APs (MUAPs) prefiltered using Wiener filtering. mean fiber concentration (MFC) jitter estimated from MUAPs. results indicate that assessment MFC may well benefit approach be EMG with accuracy comparable traditional single EMG.Third, automatic, rather than manual, detection discrimination recorded C-fiber algorithm, detects Aps reliably matched filter. Then, detected discriminated multiple hypothesis tracking combined Kalman filtering which identifies originating same C-fiber. To improve performance, amplitude estimate into algorithm. Several years use show performance algorithm excellent minimal need audit.