An Audio-Based Method for Assessing Proper Usage of Dry Powder Inhalers
Abstract
:1. Introduction
2. Methodology
2.1. Data Collection
- Actuation: inhalation using the device when hearing the “click” and indicating the correct intake of the drug.
- Inhale: inhalation using the device without releasing the drug (absence of the “click” sound) either due to pure inhalation or non-arming of the device and indicating the wrong intake of the drug.
- Button Press: sound produced when button is pressed.
- Exhale: exhaling of the patient away from the inhaler.
Data Pre-Processing
2.2. Silence Removal
2.3. Feature Extraction
2.3.1. Mel Frequency Cepstral Coefficients
2.3.2. Zero Crossing Rate
2.3.3. Root Mean Square Energy
- The signal is divided into windows.
- For each window, each sample value is squared (multiplied by itself).
- The average is obtained.
- Calculate the square root of the mean.
2.3.4. Spectral Flatness
2.3.5. Spectral Centroid
2.3.6. Spectral Roll-Off
3. Results
3.1. Classification
3.2. Post-processing
- Take as input the segments classified as Actuation or Inhale.
- Calculate the average signal width.
- Detect peaks in the first 0–8000 samples of the signal.
- Find the maximum peak.
- Perform a comparison between the maximum peak value detected in the range of the samples and the average width of the signal. If (maximum peak—average width) is greater than 90% then the event is classified as Actuation. Otherwise as Inhale.
3.3. Inference Times
4. Discussion
5. Conclusions
6. Future Work
Author Contributions
Funding
Conflicts of Interest
References
- Pocket Guide for Asthma Management and Prevention. Available online: https://ginasthma.org/wp-content/uploads/2019/04/GINA-2019-main-Pocket-Guide-wms.pdf (accessed on 10 September 2020).
- Vos, T.; Abajobir, A.A.; Abate, K.H.; Abbafati, C.; Abbas, K.M.; Abd-Allah, F.; Abdulkader, R.S.; Abdulle, A.M.; Abebo, T.A.; Abera, S.F.; et al. Global, regional, and national incidence, prevalence, and years lived with disability for 328 diseases and injuries for 195 countries, 1990–2016: A systematic analysis for the Global Burden of Disease Study 2016. Lancet 2017, 390, 1211–1259. [Google Scholar]
- Cukic, V.; Lovre, V.; Dragisic, D.; Ustamujic, A. Asthma and chronic obstructive pulmonary disease (COPD)–differences and similarities. Mater. Socio Medica 2012, 24, 100. [Google Scholar]
- Global Strategy for the Diagnosis, Management, and Prevention of Chronic Obstructive Pulmonary Disease. Available online: https:https://goldcopd.org/wp-content/uploads/2019/11/GOLD-2020-REPORT-ver1.0wms.pdf (accessed on 10 September 2020).
- Masoli, M.; Fabian, D.; Holt, S.; Beasley, R.; Global Initiative for Asthma (GINA) Program. The global burden of asthma: Executive summary of the GINA Dissemination Committee report. Allergy 2004, 59, 469–478. [Google Scholar] [PubMed]
- The Inhaler Error Steering Committee; Price, D.; Bosnic-Anticevich, S.; Briggs, A.; Chrystyn, H.; Rand, C.; Scheuch, G.; Bousquet, J. Inhaler competence in asthma: Common errors, barriers to use and recommended solutions. Respir. Med. 2013, 107, 37–46. [Google Scholar]
- Ocakli, B.; Ozmen, I.; Tunçay, E.A.; Gungor, S.; Altinoz, H.; Adiguzel, N.; Sak, Z.A.; Gungor, G.; Karakurt, Z.; Arbak, P. A comparative analysis of errors in inhaler technique among COPD versus asthma patients. Int. J. Chronic Obstr. Pulm. Dis. 2018, 13, 2941. [Google Scholar]
- Lindh, A.; Theander, K.; Arne, M.; Lisspers, K.; Lundh, L.; Sandelowsky, H.; Ställberg, B.; Westerdahl, E.; Zakrisson, A.B. Errors in inhaler use related to devices and to inhalation technique among patients with chronic obstructive pulmonary disease in primary health care. Nurs. Open 2019, 6, 1519–1527. [Google Scholar]
- Haughney, J.; Price, D.; Barnes, N.C.; Virchow, J.C.; Roche, N.; Chrystyn, H. Choosing inhaler devices for people with asthma: Current knowledge and outstanding research needs. Respir. Med. CME 2010, 3, 125–131. [Google Scholar] [CrossRef] [Green Version]
- Fink, J.B.; Rubin, B.K. Problems with inhaler use: A call for improved clinician and patient education. Respir. Care 2005, 50, 1360–1375. [Google Scholar]
- Lewis, A.; Torvinen, S.; Dekhuijzen, P.; Chrystyn, H.; Watson, A.; Blackney, M.; Plich, A. The economic burden of asthma and chronic obstructive pulmonary disease and the impact of poor inhalation technique with commonly prescribed dry powder inhalers in three European countries. BMC Health Serv. Res. 2016, 16, 251. [Google Scholar]
- Batterink, J.; Dahri, K.; Aulakh, A.; Rempel, C. Evaluation of the use of inhaled medications by hospital inpatients with chronic obstructive pulmonary disease. Can. J. Hosp. Pharm. 2012, 65, 111. [Google Scholar]
- Pritchard, J.N.; Nicholls, C. Emerging technologies for electronic monitoring of adherence, inhaler competence, and true adherence. J. Aerosol Med. Pulm. Drug Deliv. 2015, 28, 69–81. [Google Scholar] [CrossRef]
- Sulaiman, I.; Seheult, J.; Sadasivuni, N.; MacHale, E.; Killane, I.; Giannoutsos, S.; Cushen, B.; Mokoka, M.C.; Bhreathnach, A.S.; Boland, F.; et al. The impact of common inhaler errors on drug delivery: Investigating critical errors with a dry powder inhaler. J. Aerosol Med. Pulm. Drug Deliv. 2017, 30, 247–255. [Google Scholar] [CrossRef]
- Azouz, W.; Campbell, J.; Stephenson, J.; Saralaya, D.; Chrystyn, H. Improved metered dose inhaler technique when a coordination cap is used. J. Aerosol Med. Pulm. Drug Deliv. 2014, 27, 193–199. [Google Scholar] [CrossRef] [Green Version]
- Akinbami, L.J.; Simon, A.E.; Rossen, L.M. Changing trends in asthma prevalence among children. Pediatrics 2016, 137, e20152354. [Google Scholar] [CrossRef] [Green Version]
- Krishnan, V.; Diette, G.B.; Rand, C.S.; Bilderback, A.L.; Merriman, B.; Hansel, N.N.; Krishnan, J.A. Mortality in patients hospitalized for asthma exacerbations in the United States. Am. J. Respir. Crit. Care Med. 2006, 174, 633–638. [Google Scholar] [CrossRef] [Green Version]
- AAAA&I. Dry Powder Inhalers. Available online: https://www.aaaai.org/conditions-and-treatments/conditions-dictionary/dry-powder-inhalers (accessed on 29 July 2020).
- Javadzadeh, Y.; Yaqoubi, S. Therapeutic nanostructures for pulmonary drug delivery. In Nanostructures for Drug Delivery; Elsevier: Amsterdam, The Netherlands, 2017; pp. 619–638. [Google Scholar]
- Genuair, B. INN-Aclidinium Bromide. Available online: https://www.ema.europa.eu/en/documents/product-information/bretaris-genuair-epar-product-information_en.pdf (accessed on 29 July 2020).
- Berkenfeld, K.; Lamprecht, A.; McConville, J.T. Devices for dry powder drug delivery to the lung. Aaps Pharmscitech 2015, 16, 479–490. [Google Scholar] [CrossRef]
- Dal Negro, R.W.; Turco, P.; Povero, M. Patients’ usability of seven most used dry-powder inhalers in COPD. Multidiscip. Respir. Med. 2019, 14, 30. [Google Scholar] [CrossRef] [Green Version]
- Taylor, T.E.; Zigel, Y.; De Looze, C.; Sulaiman, I.; Costello, R.W.; Reilly, R.B. Advances in audio-based systems to monitor patient adherence and inhaler drug delivery. Chest 2018, 153, 710–722. [Google Scholar] [CrossRef]
- Holmes, M.S.; D’arcy, S.; Costello, R.W.; Reilly, R.B. Acoustic analysis of inhaler sounds from community-dwelling asthmatic patients for automatic assessment of adherence. IEEE J. Transl. Eng. Health Med. 2014, 2, 1–10. [Google Scholar] [CrossRef]
- Taylor, T.E.; Holmes, M.S.; Sulaiman, I.; D’Arcy, S.; Costello, R.W.; Reilly, R.B. An acoustic method to automatically detect pressurized metered dose inhaler actuations. In Proceedings of the 2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Chicago, IL, USA, 26–30 August 2014; IEEE: Piscataway, NJ, USA, 2014; pp. 4611–4614. [Google Scholar]
- Kikidis, D.; Votis, K.; Tzovaras, D. Utilizing convolution neural networks for the acoustic detection of inhaler actuations. In Proceedings of the 2015 E-Health and Bioengineering Conference (EHB), Iasi, Romania, 19–21 November 2015; IEEE: Piscataway, NJ, USA, 2015; pp. 1–4. [Google Scholar]
- Nousias, S.; Lakoumentas, J.; Lalos, A.; Kikidis, D.; Moustakas, K.; Votis, K.; Tzovaras, D. Monitoring asthma medication adherence through content based audio classification. In Proceedings of the 2016 IEEE symposium series on computational intelligence (SSCI), Athens, Greece, 6–9 December 2016; IEEE: Piscataway, NJ, USA, 2016; pp. 1–5. [Google Scholar]
- Lalos, A.S.; Lakoumentas, J.; Dimas, A.; Moustakas, K. Energy efficient monitoring of metered dose inhaler usage. J. Med Syst. 2016, 40, 285. [Google Scholar] [CrossRef]
- Butterworth, S. On the theory of filter amplifiers. Wirel. Eng. 1930, 7, 536–541. [Google Scholar]
- Giannakopoulos, T. pyaudioanalysis: An open-source python library for audio signal analysis. PLoS ONE 2015, 10, e0144610. [Google Scholar] [CrossRef]
- Zheng, F.; Zhang, G.; Song, Z. Comparison of different implementations of MFCC. J. Comput. Sci. Technol. 2001, 16, 582–589. [Google Scholar] [CrossRef]
- Chen, C.H. Signal Processing Handbook; CRC Press: Boca Raton, FL, USA, 1988; Volume 51. [Google Scholar]
- Sakhnov, K.; Verteletskaya, E.; Simak, B. Dynamical energy-based speech/silence detector for speech enhancement applications. In Proceedings of the World Congress on Engineering, Citeseer, London, UK, 1–3 July 2009; Volume 1, p. 2. [Google Scholar]
- Dubnov, S. Generalization of spectral flatness measure for non-gaussian linear processes. IEEE Signal Process. Lett. 2004, 11, 698–701. [Google Scholar] [CrossRef]
- Peeters, G. A large set of audio features for sound description (similarity and classification) in the CUIDADO project. CUIDADO IST Proj. Rep. 2004, 54, 1–25. [Google Scholar]
- Breiman, L. Random forests. Mach. Learn. 2001, 45, 5–32. [Google Scholar] [CrossRef] [Green Version]
- Tzanetakis, G.; Cook, P. Musical genre classification of audio signals. IEEE Trans. Speech Audio Process. 2002, 10, 293–302. [Google Scholar] [CrossRef]
- Cover, T.; Hart, P. Nearest neighbor pattern classification. IEEE Trans. Inf. Theory 1967, 13, 21–27. [Google Scholar] [CrossRef]
- Geurts, P.; Ernst, D.; Wehenkel, L. Extremely randomized trees. Mach. Learn. 2006, 63, 3–42. [Google Scholar] [CrossRef] [Green Version]
- Boser, B.E.; Guyon, I.M.; Vapnik, V.N. A training algorithm for optimal margin classifiers. In Proceedings of the Fifth Annual Workshop on Computational Learning Theory, Pittsburgh, PA, USA, 27–29 July 1992; pp. 144–152. [Google Scholar]
- Friedman, J.H. Greedy function approximation: A gradient boosting machine. Ann. Stat. 2001, 29, 1189–1232. [Google Scholar] [CrossRef]
No | Feature | Description |
---|---|---|
1 | Zero Crossing Rate | The rate of sign-changes of the signal during the duration of a particular frame. |
2 | Energy | The sum of the squares of the signal, normalized by the length of the frame. |
3 | Entropy of Energy | The entropy of the normalized sub-frames. |
4 | Spectral Centroid | The center of gravity of the spectrum. |
5 | Spectral Spread | The second central moment of the spectrum. |
6 | Spectral Entropy | The entropy of normalized spectral energies for a set of sub-frames. |
7 | Spectral Flux | The square difference between the normalized magnitudes of the spectrum of two consecutive frames. |
8 | Spectral Roll-off | The frequency below which 90% of the magnitude distribution of the spectrum is concentrated. |
9–21 | MFCCs | MFCCs form a cepstral representation where the frequency bands are not linear but distributed on a mel scale. |
22–33 | Chroma Vector | A 12 element representation of spectral energy. |
34 | Chroma Deviation | The standard deviation of the 12 chroma coefficients. |
Classifier | Accuracy | F1-Score | Precision | Recall |
---|---|---|---|---|
Gradient Boosting | 90.92% | 89.87% | 90.44% | 89.91% |
Extra Trees | 89.42% | 88.40% | 88.86% | 88.29% |
Random Forest | 87.22% | 86.08% | 86.20% | 86.20% |
SVM | 78.96% | 77.87% | 78.29% | 77.87% |
k-NN | 76.08% | 75.63% | 77.07% | 75.99% |
Class | F1-Score |
---|---|
Actuation | 87.12% |
Inhale | 79.35% |
Button press | 96.78% |
Exhale | 96.23% |
Class | F1-score |
---|---|
Actuation | 93.12% |
Inhale | 94.08% |
Button press | 96.67% |
Exhale | 95.56% |
Classifier | Accuracy | F1-Score | Precision | Recall |
---|---|---|---|---|
Gradient Boosting | 95.21% | 94.85% | 95.04% | 94.67% |
© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
Share and Cite
Eleftheriadou, A.-C.; Vafeiadis, A.; Lalas, A.; Votis, K.; Tzovaras, D. An Audio-Based Method for Assessing Proper Usage of Dry Powder Inhalers. Appl. Sci. 2020, 10, 6677. https://doi.org/10.3390/app10196677
Eleftheriadou A-C, Vafeiadis A, Lalas A, Votis K, Tzovaras D. An Audio-Based Method for Assessing Proper Usage of Dry Powder Inhalers. Applied Sciences. 2020; 10(19):6677. https://doi.org/10.3390/app10196677
Chicago/Turabian StyleEleftheriadou, Athina-Chara, Anastasios Vafeiadis, Antonios Lalas, Konstantinos Votis, and Dimitrios Tzovaras. 2020. "An Audio-Based Method for Assessing Proper Usage of Dry Powder Inhalers" Applied Sciences 10, no. 19: 6677. https://doi.org/10.3390/app10196677