Comparative analysis of estimation of nerve conduction velocity distribution techniques

The nerve conduction velocity is a clinically valuable indicator in the diagnosis and assessment of neuromuscular disorders and neuropathies. The conduction velocity distribution (CVD) has the potential of providing more information to help asses these pathologies. Several techniques for the estimation of nerve conduction velocity distributions have been proposed in the literature. Some of these techniques do not make use of error-free models for the electrical signal recorded from skin surface. This causes to the application of biased CVD estimators. Moreover, those that describe the evoked nerve fiber signal through physical models have had difficulties in the estimation of the electrical source required for the CVD estimation techniques proposed. Tree non-invasive CVD estimation techniques are presented in this study for nerve bundles. These estimators are based on signal models developed with the use of volume conduction theory. Compound action potentials (CAP) generated by nerve fibers with different velocities are used for these techniques. Two of these techniques (Cummins and Barker) are in time domain and have iterative solutions. Cummins solution needs a mathematical formulation to generate single fiber action potential (SFAP) for estimation of CVD. However, Barker uses CAP wave recorded close to stimulation site as a SFAP because CAP waveform is similar to SFAP in short distance. On the other hand, the third technique (Hirose) is in frequency domain and it is a deconvolution estimator. Hirose solution does not require a priori SFAP waveform for CVD. The purpose of this work is to evaluate the performance of these CVD estimators. To compare these methods, modified physiological models were created on rat tails. Ischemia was applied to affect slow conducting myelinated fibers and cold exposure was used to affect fast conducting myelinated fibers. Cummins solution is sensitive for changing in slow and fast conducting myelinated fibers in CVD but it needs an accurate and stable mathematical formulation to generate SFAP for estimation of CVD. Barker solution is sensitive to fast conducting fibers in CVD but it does not detect the changes in slow conducting myelinated fibers. Because, SFAP waveform duration used in this method does not change for different velocity classes. Hirose solution is sensitive for changing in slow and fast conducting myelinated fibers in CVD. Moreover, it also provides SFAP waveform estimation. With decreasing temperature, estimated SFAP waveform peak to peak amplitude and duration increase linearly. Also, with increasing ischemia duration estimated SFAP peak to peak amplitude increases linearly and duration does not change. Future studies performed on Hirose solution may provide detailed information on nerve action potential dynamics.