Asthma prediction and monitoring

Abstract

Abstract This literature survey paper provides a comprehensive examination of recent advancements in asthma attack prediction, with a distinct focus on the combination of machine learning (ML) and mobile health technologies. Asthma, as a complex and variable condition, necessitates personalized approaches for effective management, and this survey explores the emerging landscape of predictive tools that leverage diverse data sources. The paper reviews studies that utilize various predictors, including symptoms, physiological measures, and environmental factors, to strengthen the accuracy of predicting asthma attacks. While considerable growth has been made, the survey also highlights existing challenges such as the requirement for external validation, data privacy concerns, and the significance of larger and more representative datasets. Furthermore, the paper discusses the potential implications of these predictive models in real-world clinical practice and the ongoing efforts required to seamlessly integrate the existing asthma management strategies. As the field continues to evolve, this literature survey focuses on providing a greater understanding of the present state, challenges, and future directions in the dynamic intersection of asthma management, ML, and mobile health technologies.