A nano power CMOS tinnitus detector for a fully implantable closed-loop neurodevice

Analog signal processing offers advantages from a power consumption viewpoint. The real-time tinnitus detection method described in this paper detects tinnitus by comparing ECoG/EEG signal energies from different locations in the brain according to a tinnitus “signature”. First, the proposed strategy selects appropriate ECoG/EEG bands per channel by means of band-pass filters. Next, their extracted energies are compared to their counterparts from a different (healthy) location. Tinnitus is detected only if higher theta and gamma energies associated with lower alpha energy, in comparison to corresponding signal energies from healthy brain region, are present. To verify the detector performance, a tinnitus CMOS detector circuit has been designed to be implemented in AMIS 0.35μm technology (I3T25) and has been verified by means of simulations in Cadence using RF spectre. The final circuit operates from a 1V supply and consumes only 60nA. The applicability of the detector is verified by means of circuit simulations with real neural waveforms and is able to successfully detect tinnitus.