Electrode selection and other design considerations for a modular, portable single-channel EEG-augmented hearing aid

Abstract

In this paper, we present a design for a portable single-channel EEG-controlled hearing aid. The design has relevance for future versions of hearing aids that may make use of electroencephalography (EEG) to control the settings and signal processing on the hearing aid. The design also has relevance for auditory-based brain-computer interface (BCI) systems. Current hearing aid technology uses only acoustic information to perform noise removal algorithms and to do directional beamforming using array microphones. Recent advances in EEG and applications in auditory-based BCI have opened the door for the potential use of EEG information to automatically control the function of a hearing aid. In this paper we describe the design of a modular mobile hearing aid platform that accepts EEG inputs. We have investigated which electrodes are the most relevant for performing an EEG-based classifier of oddball sounds using tones and nonsense syllables so that eventually a minimal set of electrodes may be used in a portable auditory-based EEG BCI system. And we have described the modular design of a portable hearing aid system which enables interchange of different USB-enabled EEG electrodes, codecs, and battery supplies. The system includes a dual-microphone array for enabling adaptive beamforming noise cancelation and a Linux-based processor for performing advanced noise cancelation and compression algorithms.