Fully 3D-Printed Dry EEG Electrodes.

Electroencephalography (EEG) is used to detect brain activity by recording electrical signals across various points on the scalp. Recent technological advancement has allowed brain signals to be monitored continuously through the long-term usage of EEG wearables. However, current EEG electrodes are not able to cater to different anatomical features, lifestyles, and personal preferences, suggesting the need for customisable electrodes. Despite previous efforts to create customisable EEG electrodes through 3D printing, additional processing after printing is often needed to achieve the required electrical properties. Although fabricating EEG electrodes entirely through 3D printing with a conductive material would eliminate the need for further processing, fully 3D-printed EEG electrodes have not been seen in previous studies. In this study, we investigate the feasibility of using a low-cost setup and a conductive filament, Multi3D Electrifi, to 3D print EEG electrodes. Our results show that the contact impedance between the printed electrodes and an artificial phantom scalp is under 550 Ω, with phase change of smaller than -30∘, for all design configurations for frequencies ranging from 20 Hz to 10 kHz. In addition, the difference in contact impedance between electrodes with different numbers of pins is under 200 Ω for all test frequencies. Through a preliminary functional test that monitored the alpha signals (7-13 Hz) of a participant in eye-open and eye-closed states, we show that alpha activity can be identified using the printed electrodes. This work demonstrates that fully 3D-printed electrodes have the capability of acquiring relatively high-quality EEG signals.

Decoding of Hand Gestures from Electrocorticography with LSTM Based Deep Neural Network.

Hand gesture decoding is a key component of controlling prosthesis in the area of Brain Computer Interface (BCI). This study is concerned with classification of hand gestures, based on Electrocorticography (ECoG) recordings. Recent studies have utilized the temporal information in ECoG signals for robust hand gesture decoding. In our preliminary analysis on ECoG recordings of hand gestures, we observed different power variations in six frequency bands ranging from 4 to 200 Hz. Therefore, the current trend of including temporal information in the classifier was extended to provide equal importance to power variations in each of these frequency bands. Statistical and Principal Component Analysis (PCA) based feature reduction was implemented for each frequency band separately, and classification was performed with a Long Short-Term Memory (LSTM) based neural network to utilize both temporal and spatial information of each frequency band. The proposed architecture along with each feature reduction method was tested on ECoG recordings of five finger flexions performed by seven subjects from the publicly available 'fingerflex' dataset. An average classification accuracy of 82.4% was achieved with the statistical based channel selection method which is an improvement compared to state-of-the-art methods.