Recognition of Breathing Activity and Medication Adherence using LSTM Neural Networks

Pettas, Dionysis; Nousias, Stavros; Zacharaki, Evangelia I.; Μουστάκας, Κωνσταντίνος

doi:10.1109/bibe.2019.00176

Cited by 11 publications

(17 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…From Tables 2 and 3, it is apparent that the classification accuracy achieved by our approach does not exceed the performance of the relevant state of the art approaches. In fact, our approach performs, similarly, with the methods developed by Holmes et al [17,18], Taylor et al [24] and Pettas et al [15], but the approach of Nousias et al [13] outperforms our algorithm, mainly, for the drug and environmental noise classes. However, the utilization of a CNN architecture in the time domain allows for an implicit signal representation, that circumvents the need of additional feature extraction (e.g., in the spectral domain) and, thereby, results in significantly lower execution times.…”

Section: Comparison With Relevant Previous Worksupporting

confidence: 68%

“…Pettas et al [15] employed a deep learning based approach using the Spectrogram as a tool to develop a classifier of inhaler sounds. The Spectrogram is swept across the temporal dimension with a sliding window with length w = 15 moving at a step size equal to a single window.…”

Section: Related Workmentioning

confidence: 99%

“…Also, the annotation of the different actions was subsequently verified and completed by a trained researcher and based on visual inspection of the acquired temporal signal. In total, 360 audio files were recorded with a duration of twelve seconds each [13][14][15].…”

Section: Data Acquisitionmentioning

confidence: 99%

“…Pettas et. al [15] developed a recurrent neural network with long short term memory (LSTM), which was tested on the same dataset with this study and using the same modeling schemes, that is, SingleSubj, MultiSubj and LOSO. For the subject-specific modeling case the overall prediction accuracy was 94.75%, with higher accuracy in the prediction of breathing sounds (98%).…”

Section: Comparison With Relevant Previous Workmentioning

confidence: 99%

“…Motivated by the aforementioned open issues, this study lies on the same track as several data-driven approaches [13][14][15], presenting a method that recognizes the respiration and drug delivery phases on acoustic signals derived from pMDI usage. The main focus of this work is the investigation of acceleration aspects, namely filter pruning, scalar pruning and vector quantization, applied on convolutional neural networks (CNNs).…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Deep CNN Sparse Coding for Real Time Inhaler Sounds Classification

Ntalianis

Fakotakis

Nousias

et al. 2020

Sensors

Self Cite

View full text Add to dashboard Cite

Effective management of chronic constrictive pulmonary conditions lies in proper and timely administration of medication. As a series of studies indicates, medication adherence can effectively be monitored by successfully identifying actions performed by patients during inhaler usage. This study focuses on the recognition of inhaler audio events during usage of pressurized metered dose inhalers (pMDI). Aiming at real-time performance, we investigate deep sparse coding techniques including convolutional filter pruning, scalar pruning and vector quantization, for different convolutional neural network (CNN) architectures. The recognition performance has been assessed on three healthy subjects following both within and across subjects modeling strategies. The selected CNN architecture classified drug actuation, inhalation and exhalation events, with 100%, 92.6% and 97.9% accuracy, respectively, when assessed in a leave-one-subject-out cross-validation setting. Moreover, sparse coding of the same architecture with an increasing compression rate from 1 to 7 resulted in only a small decrease in classification accuracy (from 95.7% to 94.5%), obtained by random (subject-agnostic) cross-validation. A more thorough assessment on a larger dataset, including recordings of subjects with multiple respiratory disease manifestations, is still required in order to better evaluate the method's generalization ability and robustness.

show abstract

Section: Comparison With Relevant Previous Worksupporting

confidence: 68%

Section: Related Workmentioning

confidence: 99%

Section: Data Acquisitionmentioning

confidence: 99%

Section: Comparison With Relevant Previous Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Deep CNN Sparse Coding for Real Time Inhaler Sounds Classification

Ntalianis

Fakotakis

Nousias

et al. 2020

Sensors

Self Cite

View full text Add to dashboard Cite

show abstract

Take-A-Breath: A Novel Training Approach for Inhaler Usage Through AI-Based Self-Management and Empowerment Platform

Siopis,

Lalas,

Votis

et al. 2024

IFMBE Proceedings

View full text Add to dashboard Cite

Predicting medication adherence using ensemble learning and deep learning models with large scale healthcare data

Zalkikar²,

Liu

et al. 2021

Sci Rep

View full text Add to dashboard Cite

Clinical studies from WHO have demonstrated that only 50–70% of patients adhere properly to prescribed drug therapy. Such adherence failure can impact therapeutic efficacy for the patients in question and compromises data quality around the population-level efficacy of the drug for the indications targeted. In this study, we applied various ensemble learning and deep learning models to predict medication adherence among patients. Our contribution to this endeavour involves targeting the problem of adherence prediction for a particularly challenging class of patients who self-administer injectable medication at home. Our prediction pipeline, based on event history, comprises a connected sharps bin which aims to help patients better manage their condition and improve outcomes. In other words, the efficiency of interventions can be significantly improved by prioritizing the patients who are most likely to be non-adherent. The collected data comprising a rich event feature set may be exploited for the purposes of predicting the status of the next adherence state for individual patients. This paper reports on how this concept can be realized through an investigation using a wide range of ensemble learning and deep learning models on a real-world dataset collected from such a system. The dataset investigated comprises 342,174 historic injection disposal records collected over the course of more than 5 years. A comprehensive comparison of different models is given in this paper. Moreover, we demonstrate that the selected best performer, long short-term memory (LSTM), generalizes well by deploying it in a true future testing dataset. The proposed end-to-end pipeline is capable of predicting patient failure in adhering to their therapeutic regimen with 77.35 % accuracy (Specificity: 78.28 %, Sensitivity: 76.42%, Precision: 77.87%,F1 score: 0.7714, ROC AUC: 0.8390).

show abstract

Recognition of Breathing Activity and Medication Adherence using LSTM Neural Networks

Cited by 11 publications

References 21 publications

Deep CNN Sparse Coding for Real Time Inhaler Sounds Classification

Deep CNN Sparse Coding for Real Time Inhaler Sounds Classification

Take-A-Breath: A Novel Training Approach for Inhaler Usage Through AI-Based Self-Management and Empowerment Platform

Predicting medication adherence using ensemble learning and deep learning models with large scale healthcare data

Contact Info

Product

Resources

About