A low-power integrated circuit for adaptive detection of action potentials in noisy signals

The advent of microelectrode arrays allowing for the simultaneous recording of 100 or more neurons is leading to significant advances in science and medicine. However, the amount of data generated by these arrays presents a technical challenge if these systems are ever to be fully implanted for neuroprosthetic applications. We have developed an algorithm to perform real-time data reduction by detecting action potentials, or "spikes," embedded in a noisy signal. This algorithm is simple enough to be implemented in a mixed-signal integrated circuit consuming less than 60 muW of power. Experimental results from a chip show that the circuit is able to adaptively set a spike detection threshold above the background noise level of a signal.