Integrating Brain-Computer Interface (BCI) for Assessing Olfactory Perceptual-Ability of Human Subjects using TSK-Induced Interval Type-2 Fuzzy Classifier

Olfactory perceptual-ability in human subjects refers to the process of recognizing aromas at varying concentration levels. The aim of the paper is to introduce a novel approach that helps to evaluate perceptual–ability of human subjects with various odor concentration levels using Brain-Computer Interface (BCI) techniques. The most important non-invasive method for brain data acquisition is the Electroencephalogram (EEG). EEG signals can be used to identify brain activation regions and neural pathways involved in processing different odor stimuli. Due to the non-stationary characteristics of brain signals, the Takagi-Sugeno-Kang (TSK) induced Interval Type-2 Fuzzy logic is suitable for decoding brain signals to assess the perceptual-ability of human subjects. Three different concentration levels (High, Medium and Low) of five odor stimulus are utilized for the experiments. The analysis of the acquired brain signals using e-LORETA software confirms the involvement of the prefrontal and frontal cortex in the specified cognitive task. The experimental results demonstrate that the proposed method significantly enhance the precision of olfactory assessments compared to traditional subjective methods. A Friedman statistical test confirms that the proposed classification algorithm is superior to other classification algorithms. Moreover, in real time scenario, the recognition of olfactory stimulus and olfactory perceptual-ability assessment of human subjects is applicable for the bio-marker/ early diagnosis of Alzheimer disease and neuro-degenerative disorder.