Experimental Design of a Model between EEG Signals and Brain Regions Mapping in Anxiety Correlating Factors

Abstract

This paper applies classifying and regression machine learning algorithms within the experimental design of a model based on a holistic methodology to identify, and model EEG signals so that further stages of the experiment can map patterns within the brain in patients with anxiety. The purpose of this research is to propose forms of data analysis that could help medical specialists better diagnose and understand anxiety. This experiment methodology is based on Lean UX and the IBM Data Science Methodology. Algorithmic techniques for classification and categorization were analyzed and compared considering the projects’ requirements and constraints. Results from K Means Clustering and ID3/J48 decisions trees were compared to identify control variables and improve data quality in further iterations of the methodology alongside the model and subsequent process outline of the ongoing work.