An Effective Identification of Flavor Complaint By Adaptive Analysis of Electroencephalogram (EEG) Signal

The ability to discriminate among various foods and materials for consumption is essential for maintaining a balanced diet and lifestyle. If one's sense of taste is impaired, one's ability to take in the nutrients necessary to maintain good health is compromised. A person's reaction to a recognised tastant is typically used by doctors to identify a taste problem. Analysis of brain images captured by cutting-edge equipment like MRI can also help in the diagnosis of taste disorders. In this paper, we present an Electroencephalogram (EEG)-based technique for detecting taste dysfunction. The proposed method begins with the collection of an individual's EEG signal, then moves on to pre-processing that signal to remove any noises or disturbances, then features are extracted to make the data set more manageable, and finally a classifier is used to label the signal as coming from a healthy individual or one with a taste disorder. We were able to offer novel methods for pre-processing and feature extraction thanks to a thorough literature analysis that included a qualitative comparison of existing methods and the identification of gaps. We propose two new feature extraction algorithms-the Hilbert Huang transform with independent component analysis (HICA) and the Hilbert Huang transform principal component analysis (HPCA)-to cut down on the raw data volume of the EEG. Algorithms like ICA, DWICA, PCA, and DWPCA are used to benchmark the suggested algorithms' performance in terms of % of variance.’ As a percentage of variance, the suggested HICA and HPCA method performs better than the state-of-the-art approach.