Wireless and Wearable Auditory EEG Acquisition Hardware Using Around-The-Ear cEEGrid Electrodes.

Aside from a clinical interest in electroencephalography (EEG) measurements of real-time data with a high temporal resolution, there is a demand for acquisition systems that are operable outside the laboratory environment. In this study, we designed a wearable and low-power EEG system for multichannel EEG acquisition beyond the lab doors. Around-the-ear cEEGrid electrodes are used to capture 8 biopotential channels which are amplified by low-power precision instrumentation amplifiers and passed on to an analog-to-digital converter (ADC). An ESP32 micro-controller captures the data from the ADC and stores it on an external SD card. The proposed system is compared to a state-of-the-art EEG acquisition system (BioSemi) to assess its diagnostic power for auditory brainstem responses (ABRs). The recordings with our portable system match, and even outperform, the baseline method's specifications. Our hardware opens up new avenues for fast sampling-rate auditory EEG recordings that can be used in hearing diagnostics, damage prevention and treatment follow up.

A convolutional neural-network model of human cochlear mechanics and filter tuning for real-time applications.

Auditory models are commonly used as feature extractors for automatic speech-recognition systems or as front-ends for robotics, machine-hearing and hearing-aid applications. Although auditory models can capture the biophysical and nonlinear properties of human hearing in great detail, these biophysical models are computationally expensive and cannot be used in real-time applications. We present a hybrid approach where convolutional neural networks are combined with computational neuroscience to yield a real-time end-to-end model for human cochlear mechanics, including level-dependent filter tuning (CoNNear). The CoNNear model was trained on acoustic speech material and its performance and applicability were evaluated using (unseen) sound stimuli commonly employed in cochlear mechanics research. The CoNNear model accurately simulates human cochlear frequency selectivity and its dependence on sound intensity, an essential quality for robust speech intelligibility at negative speech-to-background-noise ratios. The CoNNear architecture is based on parallel and differentiable computations and has the power to achieve real-time human performance. These unique CoNNear features will enable the next generation of human-like machine-hearing applications.