Inhalation Motion Analysis and Visualization of Error Areas Using Two Inertial Measurement Unit Sensors

Asthma and chronic obstructive pulmonary disease (COPD) are common respiratory diseases, and the proper use of inhalers is crucial for effective symptom management. However, many patients use their inhalers incorrectly. This study proposes a method for evaluating inhaler-use behavior by employing inertial sensors to monitor patients with bronchial asthma or COPD. By augmenting an ElliptaTM inhaler with inertial measurement units, this study evaluates patient inhalation motions using the acquired motion data. Compared with conventional methods, the proposed method is less affected by external factors such as ambient sound and temperature and can be applied outside clinical settings. The evaluation process uses a linear discriminant analysis algorithm to identify key characteristic variables associated with specific inhaler-use errors, and a judgment method using these variables is proposed. A dynamic programming matching algorithm is then applied to assess correctness. Experimental results indicate that the proposed method achieves high discriminant accuracy. The proposed system enables error visualization by considering waveform similarity, unlike contemporary methods. The proposed inhalation method and accompanying dataset demonstrates a novel method for objective inhaler-use evaluation.