ABSTRACT

 The electronic tongue (E-tongue) system has emergedas a significant innovation, aiming to replicate the complexity ofhuman taste perception. In spite of the advancements in E-tonguetechnologies, two primary challenges remain to be addressed. First,evaluating the actual taste is complex due to interactions between tasteand substances, such as synergistic and suppressive effects. Second,ensuring reliable outcomes in dynamic conditions, particularly whenfaced with high deviation error data, presents a significant challenge.The present study introduces a bioinspired artificial E-tongue systemthat mimics the gustatory system by integrating multiple arrays of tastesensors to emulate taste buds in the human tongue and incorporatinga customized deep-learning algorithm for taste interpretation. Thedeveloped E-tongue system is capable of detecting four distinct tastesin a single drop of dietary compounds, such as saltiness, sourness, astringency, and sweetness, demonstrating notable reversibility andselectivity. The taste profiles of six different wines are obtained by the E-tongue system and demonstrated similarities in taste trendsbetween the E-tongue system and user reviews from online, although some disparities still exist. To mitigate these disparities, aprototype-based classifier with soft voting is devised and implemented for the artificial E-tongue system. The artificial E-tonguesystem achieved a high classification accuracy of ∼95% in distinguishing among six different wines and ∼90% accuracy even in anenvironment where more than 1/3 of the data contained errors. Moreover, by harnessing the capabilities of deep learningtechnology, a recommendation system was demonstrated to enhance the user experience.KEYWORDS: bioinspired, flexible electronics, electrochemical sensor, E-tongue, artificial Intelligence