Time-Varying Respiratory Feature Recognition and Classification for Respiration Health and Disease Screening in Children

Thorough respiratory assessments involving auscultation of breath sounds are essential for healthcare providers to accurately diagnose disease. However, parents and caregivers may lack the essential knowledge or skills to discern whether or not early signs and symptoms of respiratory disease, such as alterations in breath sounds or patterns, are problematic. In this paper, we present a novel respiration recognition and classification system for children. Young children’s breath sound signals have been collected in data sets and pre-processed as the input data. A new time-varying acoustic feature is designed and collected from the data sets. The proposed novel classifiers can recognize and classify children’s breath sounds based on the features of interest. Clinical data were used to design and verify the proposed approaches. Experiments show that the proposed approaches offer accurate results.