AI Driven Classification of Diabetes Patients’ Health Behavior

This study aims to predict the health-seeking behavior among diabetic patients based on their Emotional Intelligence and Health Locus of Control with the help of artificial intelligence and Internet of Things. Health-seeking behavior is a critical determinant of healthcare outcomes. It is influenced by various factors like sociodemographic characteristics, emotional intelligence, and internal/external locus of control. While the interplay between these variables has been explored in prior research, this study uniquely combines them to classify patients into high or low health-seeking behavior categories. The research is exploratory in nature. 600 diabetic patients in Sikkim, India were surveyed with the help of Brief Emotional Intelligence Scale-10 and the Multidimensional Health Locus of Control Scale. Machine learning algorithm - neural network was used for classification of the patients. Results indicate a classification accuracy exceeding 90%, with the area under the ROC curve surpassing 0.9 which demonstrates the effectiveness of model. These findings underscore the potential of artificial intelligence and the Internet of Things in predicting health-seeking behavior. It can be used by healthcare providers in tailoring interventions for improved health outcomes. This novel approach helps to bridge the gaps in healthcare accessibility and fosters advancements in patient-centered care strategies.