Objective Assessment of Patient Inhaler User Technique Using an Audio-Based Classification Approach

Abstract

Many patients make critical user technique errors when using pressurised metered dose inhalers (pMDIs) which reduce the clinical efficacy of respiratory medication. Such critical errors include poor actuation coordination (poor timing of medication release during inhalation) and inhaling too fast (peak inspiratory flow rate over 90 L/min). Here, we present a novel audio-based method that objectively assesses patient pMDI user technique. The Inhaler Compliance Assessment device was employed to record inhaler audio signals from 62 respiratory patients as they used a pMDI with an In-Check Flo-Tone device attached to the inhaler mouthpiece. Using a quadratic discriminant analysis approach, the audio-based method generated a total frame-by-frame accuracy of 88.2% in classifying sound events (actuation, inhalation and exhalation). The audio-based method estimated the peak inspiratory flow rate and volume of inhalations with an accuracy of 88.2% and 83.94% respectively. It was detected that 89% of patients made at least one critical user technique error even after tuition from an expert clinical reviewer. This method provides a more clinically accurate assessment of patient inhaler user technique than standard checklist methods.

Hold the device against the skin: the impact of injection duration on user’s force for handheld autoinjectors

“…Although prior research on pulmonary drug delivery has routinely used sensor-based approaches to characterize patients' behaviors [e.g. [60][61][62][63][64], such approaches did not stimulate research on subcutaneous self-injections. To the best of the authors' knowledge, the only exception is the study published by Xiao and colleagues [44] on measurements of needle displacements derived from device motion to assess patients' completion of injections.…”

Investigating the validity and reliability of the traditional checklist for assessment of specific inhalation techniques with pressurized metered-dose inhalers using the Aerosol Inhalation Monitor

“…The agreement was poor between the checklist and AIM. In the study of Taylor et al [32], the fair-to-moderate agreement (according to the Cohen's κ statistic) between the subjective visual checklist assessment and the Inhaler Compliance Assessment audio recording device (κ = 0.49 for actuation coordination assessment and κ = 0.36 for PIFR assessment) highlighted the potential inaccuracy of the checklist method in assessing patient inhaler user technique.…”

Deep CNN Sparse Coding for Real Time Inhaler Sounds Classification

“…A year later, Holmes et al [18,19], also, developed an algorithm that recognizes blister events and breath events (with an accuracy of 92.1%) and separates inhalations from exhalations (with an accuracy of more than 90%). Later, Taylor et al developed two main algorithms for blister detection [26,37] based on Quadratic Discriminant Analysis and ANN, and achieved an accuracy of 88.2% and 65.6%, respectively. Nousias et al in Reference [13] presented a comparative study between Random Forest, ADABoost, Support Vector Machines and Gaussian Mixture Models, reaching the conclusion that RF and GMM yield a 97% to 98% classification accuracy on the examined dataset, when utilizing MFCC, Spectrogram and Cepstrogram features.…”