The role of artificial intelligence in the differential diagnosis of wheezing symptoms in children

Abstract

The incidence of pediatric wheeze is extremely high. Poor control of wheeze in young children affects lung function in adulthood and is closely associated with the occurrence of chronic obstructive pulmonary disease. Substantial efforts worldwide have been aimed at developing methods to identify the etiology of wheezing symptoms as early as possible to aid in early management strategies. However, the diagnosis of childhood wheeze relies heavily on the clinical experience of pediatricians, most of whom lack sufficient training to accurately diagnose children with wheezing symptoms. Artificial intelligence is an approach that may improve general pediatricians’ diagnostic ability for wheezing symptoms by identifying patterns and trends from large and complex clinical datasets. However, few studies have used artificial intelligence to diagnose wheeze in children. Therefore, this review aims to comprehensively assess these studies in this field, analyze their interpretability and limitations, and explore and discuss future research directions in real-world clinical applications.