AI-ENABLED MULTIMODAL RISK ASSESSMENT FOR EARLY DETECTION AND PREVENTION OF PEDIATRIC ASTHMA AND ASTHMA EXACERBATIONS

Abstract

Asthma is a chronic condition that increases the risk of symptoms and health emergencies for children and raises the economic burden on families and societies. In fact, there is clear evidence to show that children with asthma experience significantly worse quality of life than their peers without asthma. Children with poorly controlled asthma are at risk for frequent, unpredictable asthma attacks. Genes, the environment, the healthcare system, the family, the child and lifestyle, in addition to behavior, interact in a complex way and make it difficult to predict asthma development and attacks. The traditional systems and methods for assessment and management of asthma attacks are inadequate for the prevention of pediatric asthma attacks. The main objective of the study was to design an AI-based, risk assessment and management system for pediatric asthma. The system is based on integrating different sources of information (clinical, environmental, physiological, behavioral) and data (time series, longitudinal, multimodal) and using innovative Artificial Intelligence/Machine Learning and Deep Learning techniques for interpreting data and discovering unchartered knowledge. The system aims to take a data-driven approach to detecting and managing pediatric asthma.