Acoustic Monitoring for Estimating Lung Deposition in Dry Powder Inhalers

This Letter presents a correlation-based and signal-conversion framework that extends the measurable functionality of an acoustic inhaler sensor, enabling quantitative estimation of lung deposition during real-world dry powder inhaler (DPI) use. A microelectromechanical systems (MEMS) microphone with a passive dual-slit acoustic filter captures inhalation sounds, which are processed through feature-based regression for subject-independent flow reconstruction. The reconstructed flow shows strong agreement with a reference flow meter ($R^{2} = {0.846}$) and is further correlated with Next Generation Impactor (NGI) measured lung dose ($R^{2} = 0.828$). By linking inhalation acoustics—originally used for qualitative detection of inhalation correctness—to quantitative estimates of lung deposition, the proposed method extends the functionality of the acoustic inhaler sensor to provide physiological measurement capability, broadening its applicability for compact and non-invasive respiratory assessment.