Author(s): Sun, YK (Sun, Yike); Shen, AR (Shen, Anruo); Sun, JN (Sun, Jingnan); Du, CL (Du, Chenlin); Chen, XG (Chen, Xiaogang); Wang, YJ (Wang, Yijun); Pei, WH (Pei, Weihua); Gao, XR (Gao, Xiaorong)

Source: IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING Volume: 30 Pages: 2042-2051 DOI: 10.1109/TNSRE.2022.3192543 Published: 2022

Abstract: The research on non-invasive BCI is nowadays hitting the bottleneck due to the humble quality of scalp EEG signals. Whereas invasive solutions that offer higher signal quality in contrast are suffocated in their spreading because of the potential surgical complication and health risks caused by electrode implantation. Therefore, it puts forward a necessity to explore a scheme that could both collect high-quality EEG signals and guarantee high-level operation safety.This study proposed a Minimally Invasive Local-skull Electrophysiological Modification method to improve scalp EEG signals qualities at specific brain regions. Six eight-month-old SD rats were used for in vivo verification experiment. A hole with a diameter of about 500 micrometers was drilled in the skull above the visual cortex of rats. Significant changes in rsEEG and SSVEP signals before and after modification were observed. After modification, the skull impedance of rats decreases by about 84 %, the average maximum bandwidth of rsEEG increase by 57 %, and the broadband SNR of SSVEP is increased by 5.13 dB. The time of piezoelectric drilling operation is strictly controlled under 30 seconds for each rat to prevent possible brain damage from overheating. Compared with traditional invasive procedures such as ECoG, Minimally Invasive Local-skull Electrophysiological Modification operation time is shorter and no electrode implantation is needed while it remarkably boosts the scalp EEG signal quality. This technical solution has the potential to replace the use of ECoG in certain application scenarios and further invigorate studies in the field of scalp EEG in the future.

Accession Number: WOS:000831114400004

PubMed ID: 35857723

Author Identifiers:

Author        Web of Science ResearcherID        ORCID Number

Chen, Xiaogang                  0000-0002-5334-1728

Sun, Yike                  0000-0001-5970-3876

ISSN: 1534-4320

eISSN: 1558-0210

Full Text: https://ieeexplore.ieee.org/document/9833539