Brain Computer Interface / BCI / Neural Control Int. / NCI / Mind Machine Int. / MMI / Direct Neural Int. / DNI / Brain Machine Int. / BMI
Marzie Alirezaei Alavijeh; Ali Maleki
Volume 16, Issue 1 , May 2022, , Pages 1-9
Abstract
Nowadays, brain-computer interface system based on steady-state visual evoked potentials is increased due to advantages such as accepted accuracy and minimal need for user training. Despite these benefits, the unwanted noise that affects SSVEP is one of the issues that can reduce the efficiency of such ...
Read More
Nowadays, brain-computer interface system based on steady-state visual evoked potentials is increased due to advantages such as accepted accuracy and minimal need for user training. Despite these benefits, the unwanted noise that affects SSVEP is one of the issues that can reduce the efficiency of such systems. This paper uses the EMD algorithm in the initial phase and CCA or LASSO for the recognition of the stimulation frequency. In the first step, the EMD algorithm is applied so that non-stationary SSVEP signal breaks into oscillating functions and meaningful information are extracted. Among the IMFs obtained from the EMD method, only IMFs whose amplitude of the frequency spectrum in the frequency ranges corresponding to the excitation is higher were selected. With this selection, noisy signals and unprofitable information can be omitted. In the proposed method, two CCA and LASSO diagnostic methods were performed on the sum of selected signals to identify the frequency of stimulation. The simulation results show the recognition accuracy of 81.76% and 82.26% for the proposed method EMD-CCA and EMD-LASSO, respectively. While detection accuracy is 78.10% and 78.72% for conventional methods of CCA and LASSO.
Human Computer Interaction / HCI
Mohsen Keshtkar; Amin Mahnam; Pegah Poladian
Volume 10, Issue 4 , January 2017, , Pages 279-290
Abstract
Steady State Visual Evoked Potentials (SSVEP) have been widely used in development of Brain Computer Interfaces (BCI). However, it is still a research challenge to have visual stimuli which provide strong SSVEP response while produce little eye fatigue. In this study, rectangular, sinosoidal, sawtooth ...
Read More
Steady State Visual Evoked Potentials (SSVEP) have been widely used in development of Brain Computer Interfaces (BCI). However, it is still a research challenge to have visual stimuli which provide strong SSVEP response while produce little eye fatigue. In this study, rectangular, sinosoidal, sawtooth waveforms applied to a LED were compared with sum of two sinusoidals and a frequency modulated waveform to determine the most appropriate visual stimulus for realization of a BCI system. Moreover, circular, ring and anti-phase two rectangular flickers were generated by Cogent toolbox on a laptop screen and compared. Experiments were performed on 12 participants to determine the SSVEP response and eye fatigue corresponding to each of these visual stimuli. Experiments with the waveforms demonstrate that sum of two sine waves generated significantly lower SSVEP amplitude, but the responses for other four waveforms were not significantly different. On the other hand, the frequency modulated waveform resulted in the least eye fatigue significantly lower from other waveforms. Therefore, considering both criteria, frequency modulated waveform can provide superor performance in a BCI system with an average response of 17.3 pV2 and 1.58 fatigue level in a 1-4 fatigue scale. Experiments with visual stimuli on LCD showed that circular stimuli provided highest and anti-phase rectangular the lowest response. But all of them produced high levels of eye fatigue. Although, Circular stimuli had the highest power (26.7pV2) but due to its related high eye fatigue (3.8) it is not recommended for practical applications. In conclusion it is recommended to use frequency modulated visual stimuli for development of practical BCI systems to satisfy both strong response and low eye fatigue criteria.
Bioelectrics
Marzieh Alirezaei Alavijeh; Ali Maleki
Volume 10, Issue 2 , August 2016, , Pages 187-196
Abstract
Brain-computer interface system based on Steady-state visual evoked potentials is taken into consideration due to advantages such as simplicity of installation and use of the system, enough accurate and acceptable Information transfer rate. In addition to these benefits, short processing time is also ...
Read More
Brain-computer interface system based on Steady-state visual evoked potentials is taken into consideration due to advantages such as simplicity of installation and use of the system, enough accurate and acceptable Information transfer rate. In addition to these benefits, short processing time is also an important criterion to have a system that is applicable in real life and have the ability to use online. In this paper, a method based on standard CCA have been present for recognition of stimulus frequency. The proposed method is performed in two stages, offline and online. In the offline stage, the standard CCA is applied to the SSVEP and sin-cos reference signals. After that, template signals are constructed using weights that generate maximum correlation. In online stage, cross correlation between test signal and each template signals are calculated and the stimulus frequency is recognized. The greater accuracy of frequency recognition and less calculation time at the same time are shown by stimulation result.