CHEWING MUSCLE DATASET

The efficiency of the mastication process in humansis closely linked to various physiological and psychological conditions.Inadequate chewing or inefficient muscle activation patternsin the muscles of mastication can serve as early indicatorsof potential health issues. Early detection of such inefficienciesenables healthcare professionals to identify underlying causes orsignals of other diseases. In this research, we explore a novelapplication of AI to analyze muscle efficiency during chewing,leveraging machine learning techniques to detect inefficiencies.With prior consent, we collected electromyography (EMG) datafrom 11 healthy subjects, four of whom had dental ailments thataffected their chewing efficiency. Rather than labeling all theirmasticatory muscles as inefficient, we focused on analyzing specificsignal properties to determine which muscles were impactedby inefficiencies. For this purpose, we developed and trained fourseparate classifier models, each corresponding to a specific muscleinvolved in mastication. Each model generates predictions regardingmuscle efficiency, and the results are further interpretedby identifying the top three contributing factors influencingthose predictions. These insights allow us to provide suggestionsregarding potential causes of chewing inefficiency. Our approachachieved a maximum classification accuracy of 95.23 in detectingmuscle efficiency, demonstrating the effectiveness of machinelearning in this domain. This research marks a significant steptoward using AI to support early diagnosis and personalizedtreatment for chewing-related issues.