P1000 Induced Brain Signal Analysis for Assessing Subjective Pain Sensitivity using Type-2 Fuzzy Classifier

This paper intends to develop a novel methodology that helps to determine the variation of pain perception across various individuals using EEG signal analysis. Three types of touch stimuli: heat, bristles and pinch with varying intensity levels are utilized for the experiment. The brain signals acquired are analyzed using eLORETA software that confirms the involvement of frontal and parietal lobes for this cognitive activity. Additionally, frequency analysis undertaken infers the participation of alpha and theta bands for the said task. The signals are further evaluated to inspect the existence of any Event Related Potential (ERP) signal. A unique and notable ERP signal has been found when a subject finds the perceived stimuli to be painful. However, no relevant ERP component is generated when the subject finds the presented stimuli to be completely painless. A novel Interval Type-2 fuzzy classifier has been designed to classify these two distinct conditions (painful and non-painful). Performance analysis undertaken confirms the superlative behaviour of the proposed classifier with respect to other standard ones. Moreover, statistical evaluation also assures the superior performance of the proposed classifier model. Hence, this method can act as a neuronal marker to detect an individual’s pain sensitivity that can be used to diagnose and treat various neurological disorders and chronic pain based diseases.