An Automated Sleep Stage Classification for Healthcare Monitoring by using Single Channel EEG Signal

It is well-established that biomedical signals convey crucial data regarding the functioning of living systems. The physiological and clinical information included in these signals can be improved with adequate processing. Modern qualitative and quantitative analyses of physiological systems and events rely on digital signal processing and pattern recognition methods. Analysis and interpretation of a medical practitioner's signal carry the weight of the analyst's knowledge and expertise, yet such analysis is inherently subjective. If done logically, computer analysis of biomedical information might provide credibility to the expert's interpretation by providing an objective second opinion. Furthermore, it allows for enhanced diagnosis and online monitoring of critically ill patients. The current research intends to develop effective methods for utilizing sleep-monitoring health gadgets. When it comes to handling complex classification or pattern recognition issues, the Support Vector Machine (SVM) is the instrument of choice. In this article, we focus on using support vector machines (SVMs) to identify and categorize apnea. When compared to other methods of categorization, such as sophisticated statistical approaches, SVM performed better. Both an adaptive classification model and a novel approach to merging the decisions of ensemble-based classification models are proposed in the work. The current method relies on an ensemble classifier system and a huge number of features, making it both effective and trustworthy.