#35 Spectral Analysis of Wheezing Children's Breath Sounds at the Children's Emergency Depertment
Matthew Kochan, University of Manitoba; Allan Becker, Children’s Hospital Research Institute of Manitoba; Saif Huq, Pediatrics and Child Health; Scott Sawyer, Children’s Hospital Research Institute of Manitoba
Asthma is a chronic, obstructive, inflammatory lung disease that affects 13% of Canadian children. It is the most common chronic disease in childhood seen in the emergency department. Asthma typically presents with coughing and wheezing, however parental reporting of a wheeze is difficult to interpret and inconsistent. As a result, diagnosis of asthma may be delayed or made prematurely. To address this issue a clearer definition of wheezing is needed for the lay audience, along with better education regarding breath sounds and respiratory disease.
In this study breath sound recordings from 15 healthy children and 15 wheezing children presenting to the Children’s Hospital Emergency Department were digitally recorded and analyzed. Children aged two months old to eight years old were recorded using a custom stethoscope connected to a mobile phone. Analysis was based on a Fast Fourier Transform of the spectrogram of the sound to create power spectrum density (PSD) curves for the breath sounds. The PSDs were used to analyze differences in amplitude and frequency throughout the breath sounds.
The results identify differences in peak inspiratory and expiratory power, and different trends in the power and frequency in expiratory breath sounds that can be used to distinguish wheezing from normal breath sounds in a pediatric population. Specifically, the slope in the inspiratory PSD curve from 200Hz to the peak inspiratory power and the slope in the expiratory PSD curve from 400-700Hz show significant differences between normal and wheezing breath sounds. In addition, the angle of the peak in the expiratory curve is significantly more acute in the wheezing children.
The results of this study can help better define wheezing for a lay audience and could be incorporated in mobile application that can assist families and health care professionals with identifying wheezing and asthma.