Quality assessment and feedback of Smart Device Microphone Spirometry executed by children

TLDR: The method proposed was able to correctly classify the microphone spirometry with respect to admissible minimum of effort with an accuracy of 86% (specificity 87% and sensitivity 86%) and can be used to provide immediate feedback of the correct execution of the maneuver, improving the clinical value and utility of this self-monitoring tool.

Abstract: Smart device microphone spirometry, based on the audio recording of forced expiratory maneuver (FEM), can be a simple, ubiquitous and easy tool for patients to self-monitor their asthma. Automatic validity assessment is crucial to guarantee that the global effort of the FEM fulfil the admissible minimum or if the maneuver needs to be repeated. In this work an automatic method to classify the sounds from FEM with respect to global effort was developed and evaluated using data from 54 children (5-10 years). The method proposed was able to correctly classify the microphone spirometry with respect to admissible minimum of effort with an accuracy of 86% (specificity 87% and sensitivity 86%). This method can be used to provide immediate feedback of the correct execution of the maneuver, improving the clinical value and utility of this self-monitoring tool.