Recognition of Breathing Activity and Medication Adherence using LSTM Neural Networks

Obstructive inflammatory pulmonary diseases are life-long conditions of the airways affecting millions worldwide. A crucial step towards effective self-management is the adherence of the patients to their medication. Accurate detection of pressurised metered dose inhaler audio events can significantly improve medication adherence facilitating more meaningful interventions by medical personnel. Towards this direction, this work presents a data-driven approach for monitoring pressurised metered dose inhaler medication adherence employing recurrent neural networks with long short term memory (LSTM) units with spectrogram features. Evaluation studies took place with intra-subject and inter-subject settings, demonstrating that our LSTM-based approach yields higher accuracy compared to classical machine learning methods reaching a prediction accuracy ranging from 92% to 94% taking into account also samples comprised of a mixture of patterns belonging to more than one of the predefined classes.

[1]  D. Fabian,et al.  The global burden of asthma: executive summary of the GINA Dissemination Committee Report , 2004, Allergy.

[2]  Richard B Reilly,et al.  Estimation of inhalation flow profile using audio-based methods to assess inhaler medication adherence , 2018, PloS one.

[3]  Yang Bin,et al.  Audio Events Detection and classification using extended R-FCN Approach , 2017, DCASE.

[4]  Liu Liu,et al.  Learning Effective Binary Descriptors via Cross Entropy , 2017, 2017 IEEE Winter Conference on Applications of Computer Vision (WACV).

[5]  Dimitrios Tzovaras,et al.  An mHealth System for Monitoring Medication Adherence in Obstructive Respiratory Diseases Using Content Based Audio Classification , 2018, IEEE Access.

[6]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[7]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[8]  Yaniv Zigel,et al.  Advances in Audio-Based Systems to Monitor Patient Adherence and Inhaler Drug Delivery. , 2017, Chest.

[9]  Yaniv Zigel,et al.  Objective Assessment of Patient Inhaler User Technique Using an Audio-Based Classification Approach , 2018, Scientific Reports.

[10]  Geoffrey E. Hinton,et al.  Speech recognition with deep recurrent neural networks , 2013, 2013 IEEE International Conference on Acoustics, Speech and Signal Processing.

[11]  Alan E. Simon,et al.  Changing Trends in Asthma Prevalence Among Children , 2016, Pediatrics.

[12]  Richard Beasley,et al.  Validation of a metered dose inhaler electronic monitoring device: implications for asthma clinical trial use , 2016, BMJ Open Respiratory Research.

[13]  Mathieu Lagrange,et al.  Detection and Classification of Acoustic Scenes and Events: Outcome of the DCASE 2016 Challenge , 2018, IEEE/ACM Transactions on Audio, Speech, and Language Processing.

[14]  Andrey Temko,et al.  CLEAR Evaluation of Acoustic Event Detection and Classification Systems , 2006, CLEAR.

[15]  Sarah Sharples,et al.  Electronic monitoring of adherence to inhaled medication in asthma , 2014 .

[16]  B. Aggarwal,et al.  Use of pressurized metered dose inhalers in patients with chronic obstructive pulmonary disease: review of evidence , 2014, Expert review of respiratory medicine.

[17]  Richard B. Reilly,et al.  An acoustic method to automatically detect pressurized metered dose inhaler actuations , 2014, 2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[18]  Vidya Krishnan,et al.  Mortality in patients hospitalized for asthma exacerbations in the United States. , 2006, American journal of respiratory and critical care medicine.

[19]  Ian McLoughlin,et al.  What makes audio event detection harder than classification? , 2016, 2017 25th European Signal Processing Conference (EUSIPCO).

[20]  Richard B. Reilly,et al.  Automatic identification and accurate temporal detection of inhalations in asthma inhaler recordings , 2012, 2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[21]  Ian McLoughlin,et al.  Continuous robust sound event classification using time-frequency features and deep learning , 2017, PloS one.

[22]  Christopher E Brightling,et al.  The relationship between clinical outcomes and medication adherence in difficult-to-control asthma , 2012, Thorax.

[23]  R. Reilly,et al.  Acoustic Analysis of Inhaler Sounds From Community-Dwelling Asthmatic Patients for Automatic Assessment of Adherence , 2014, IEEE Journal of Translational Engineering in Health and Medicine.

[24]  Dan Stowell,et al.  Deep Learning for Audio Event Detection and Tagging on Low-Resource Datasets , 2018, Applied Sciences.

[25]  Alex Sherstinsky,et al.  Fundamentals of Recurrent Neural Network (RNN) and Long Short-Term Memory (LSTM) Network , 2018, Physica D: Nonlinear Phenomena.

[26]  C. Gillette,et al.  Inhaler Technique in Children With Asthma: A Systematic Review. , 2016, Academic pediatrics.