Identification of spectral landmarks on the single fiber action potential waveform for unit tracking

Highly selective indwelling neuroprosthetic interfaces are able to detect minute electrical signals from the nerve fibers in its vicinity as single fiber action potential (SFAP) waveforms. In the peripheral nervous system, each SFAP has a unique shape resulting from each fiber's unique microelectrical interaction between itself, it's mileau and the recording electrode. Two specific factors, the nerve fiber's position relative to electrode and its conduction velocity, greatly influence the SFAP shape. Thus, we hypothesize that these two factors embedded in the SFAP could be ultimately extracted and used as a means to non-destructively assess the health of each nerve-electrode couple in chronically implanted neural interfaces. Here, we present a method for estimation of these two factors through quantitative spectral analysis of the SFAP shape in intrafascicular recordings from the rabbit sciatic nerve.