Advanced region specific neurofeedback system for depression

Abstract

A number of EEG based computer-aided systems for clinical use have been presented in recent years. EEG-Neurofeedback based operant conditioning is one of them. There has been insufficient evidence of neurofeedback for the depression symptoms. The performance of such systems is limited as EEG is contaminated by various noise sources and its spatial resolution is poor due to the inverse problem. In the present work, a series of experiments have been conducted for EEG denoising using the wavelet domain, independent component analysis, and empirical mode decomposition domains. And, a methodology for hybrid artifact removal based on a combination of variational mode decomposition with the wavelet domain using detrended fluctuation analysis has been proposed. Also, to improve the EEG inverse problem, a new method of EEG source localization using variational mode decomposition and sLORETA inverse model is proposed. The proposed algorithm has shown efficacy in terms of various performance parameters for denoising and inverse source localization than conventional approaches for EEG signals of depression. It can be said that the current research work is very promising for EEG based neurofeedback systems. The user may experience a wider range of feedback features resulting in better results. Such a system may offer an effective solution for clinicians as a crucial stage of EEG pre-processing in automated depression detection systems. newline