Abstract
Understanding the brain is important in the fields of science, medicine, and engineering. A promising approach to better understand the brain is through computing models. These models were adjusted to reproduce data collected from the brain. One of the most commonly used types of data in neuroscience comes from electroencephalography (EEG), which records the tiny voltages generated when neurons in the brain are activated. In this study, we propose a model based on complex networks of weakly connected dynamical systems (Hindmarsh–Rose neurons or Kuramoto oscillators), set to operate in a dynamic regime recognized as Collective Almost Synchronization (CAS). Our model not only successfully reproduces EEG data from both healthy and epileptic EEG signals, but it also predicts EEG features, the Hurst exponent, and the power spectrum. The proposed model is able to forecast EEG signals 5.76 s in the future. The average forecasting error was 9.22%. The random Kuramoto model produced the outstanding result for forecasting seizure EEG with an error of 11.21%.
Highlights
The proposed model was applied to predict the EEG signals of five datasets
According to a previous work[20], the output function constructed from an average of weighted (“trained”) time series collected from randomly selected neurons in a network operating in the CAS regime can well reproduce EEG signals
We found that the mean absolute error (MAE) ratio values obtained from the different datasets do not differ much for the EEG signals predicted by our four network models
Summary
The proposed model was applied to predict the EEG signals of five datasets. The model was evaluated by comparing the experimental EEG signal for the “test data” time windows (the last 1000 data points) with the predicted EEG from the model. The regression model was used to forecasting the EEG, this process is an algebraic one. Each EEG signal is only predicted using its own information. Regarding the used data set, 2000 time points for calculating the coefficient is enough to avoid the over-fitting and give the stable results. To further evaluate the performance of our model, we compared the Hurst exponent (related to the long-range correlation) and the power spectrum of the experimental EEG dataset with that obtained from our model, considering four configurations: random HR, small-world HR, random Kuramoto, and small-world Kuramoto models
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
