Real-time decoding of bladder pressure from pelvic nerve activity

Abstract

Real time algorithms for decoding physiological signals from peripheral nerve recordings form an important component of closed loop bioelectronic medicine (electroceutical) systems. As a feasibility demonstration, we considered the problem of decoding bladder pressure from pelvic nerve electroneurograms. We extracted power spectral density of the nerve signal across a band optimised for Shannon Mutual Information, followed by linearization via piece-wise linear regression, and finally decoded signal reconstruction through optimal linear filtering. We demonstrate robust and effective reconstruction of bladder pressure, both prior to and following pharmacological manipulation.