Validation of the EEG signal of the URGOnight neurofeedback device, associated with a new SMR detection method

Abstract

AbstractSensorimotor (SMR) neurofeedback is a promising therapy for several health disorders but is still not widely used due to the high cost of the equipment. URGOnight offers a low-cost solution to democratize these therapies by providing an at-home EEG headband with dry electrodes connected to a mobile application. The first aim of this study is both to validate the URGOnight EEG signal and to compare it to Enobio-20, a medical grade EEG device. The second aim of the study is to propose a new method to detect SMR rhythm based on its oscillatory properties and discriminate it from alpha oscillations.In our study, we compared the URGOnight headband EEG signal (C3/C4) to Enobio-20 (CP3/CP4), placed on subjects simultaneously equipped with the two headbands. All subjects (n=33) performed a dual blocking task inspired by Kulhman (1978) based on the blocking effect of movement and eyes opening on SMR and alpha respectively. This task was followed by SSVEP stimulations to evaluate the frequency response of the two EEG devices. The performance of the EEG headbands was statistically identical for most of the characteristics of the EEG signal, including the frequency response to SSVEP (from 4Hz to 20Hz). The main difference was a larger amplitude in the 8-15Hz due to the location of the reference in URGOnight that did not impair the detection of both alpha and SMR.In addition, we show that our new method allows to discriminate alpha and SMR rhythms based on their oscillatory properties with a single recording site (C3/C4). The method is fast enough to be used in real time. We show that the detected SMR rhythm is modulated by movement as opposed to the 12-15Hz frequency band often used as indicator of SMR in most neurofeedback studies.Altogether, our results validate the quality of the EEG recordings obtained with URGOnight since it gives similar results as the one obtained with Enobio-20, a validated EEG medical grade system. In addition, we provide a new method allowing the identification and the separation of the alpha and SMR with a single recording site C3/C4. This method opens up a new research lead to improve SMR neurofeedback efficiency and thus of its clinical possibilities by focusing on the reinforcement of the SMR oscillation strictly speaking.Highlights-Validation of the URGOnight EEG device suitable for neurofeedback at home-New method for the detection and the discrimination of alpha rhythm and SMR rhythm with a small number of recording sites-The oscillatory activity related to the SMR displays different properties compared to the 12-15Hz frequency band.-Description of a full validation procedure for wireless EEG devices usable at home for neurofeedback-Comparison of the signal of URGOnight (dry electrodes) with a wet electrode EEG device