Determination of nerve conduction velocity distribution from sampled compound action potential signals

The sampled compound action potential (CAP) data sequence was expressed as the circular convolution of the delay sequence and the sampled single fiber action potential (SFAP) data sequence. An algorithm, based on Hirose's method (1986) was then developed to separate the delay sequence from the sampled CAP data sequence, and the nerve conduction velocity distribution (NCVD) was consequently calculated from the delay sequence. The NCVD was found to be the product of the amplitude of the SFAP and the number of fibers. Simulations show that the estimated results were in good agreement with the calculated results. Experiments were performed on ten sciatic nerves from five bullfrogs (Rana pipens) using two independent variables: interelectrode distance and stimulus current strength. The results estimated from CAP's recorded under each condition reflect the corresponding feature of NCVD of the condition. The advantage of the technique is to provide detailed information about both slow and fast conducting fibers. This technique also offers the possibility to directly calculate the nerve fiber diameter distribution from the sampled CAP data sequences.