Detection of Obstructive Sleep Apnoea Using Features Extracted From Segmented Time-Series ECG Signals With a One Dimensional Convolutional Neural Network

This paper reports on ongoing research, which aims to prove that features of Obstructed Sleep Apnoea (OSA) can be automatically identified from single-lead electrocardiogram (ECG) signals using a One-Dimensional Convolutional Neural Network (1DCNN) model. The 1DCNN is also compared against other mac...

Full description

Bibliographic Details
Main Authors: Steven Thompson, Denis Reilly, Paul Fergus, Carl Chalmers
Format: Article
Language:English
Published: IEEE 2024-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10373019/
_version_ 1797365943336173568
author Steven Thompson
Denis Reilly
Paul Fergus
Carl Chalmers
author_facet Steven Thompson
Denis Reilly
Paul Fergus
Carl Chalmers
author_sort Steven Thompson
collection DOAJ
description This paper reports on ongoing research, which aims to prove that features of Obstructed Sleep Apnoea (OSA) can be automatically identified from single-lead electrocardiogram (ECG) signals using a One-Dimensional Convolutional Neural Network (1DCNN) model. The 1DCNN is also compared against other machine learning (ML) classifier models, namely Support Vector Machine (SVM) and Random Forest Classifier (RFC). The 1DCNN architecture consists of 4 major parts, a Convolutional Layer, a Flattened Dense Layer, a Max Pooling Layer and a Fully Connected Multilayer Perceptron (MLP), with 1 Hidden Layer and a SoftMax output. The model repeatedly learns how to better extract prominent features from one-dimensional data and map it to the MLP for increased prediction. Training and validation are achieved using pre-processed time-series ECG signals captured from 35 ECG recordings. Using our unique windowing strategy, the data is shaped into 5 datasets of different window sizes. A total of 15 models (5 for each group, 1DCNNs, RFCs, SVMs) were evaluated using various metrics, with each being run over numerous experiments. Results show the 1DCNN-500 model delivered the greatest degree of accuracy and rapidity in comparison to the best producing RFC and SVM classifiers. 1DCNN-500 (Sensitivity 0.9743, Specificity 0.9708, Accuracy 0.9699); RFC-500 (Sensitivity/Recall (0) 0.90 / (1) 0.94, Precision (0) 0.94 / (1) 0.90, Accuracy 0.91); SVM-500 (Sensitivity (0) 0.94 / (1) 0.50, Precision (0) 0.65 / (1) 0.90, Accuracy 0.72). The model presents a novel approach that could provide support mechanisms in clinical practice to promptly diagnose patients suffering from OSA.
first_indexed 2024-03-08T16:57:10Z
format Article
id doaj.art-b717adb2e6d3423a8a23deaf71cea4ff
institution Directory Open Access Journal
issn 2169-3536
language English
last_indexed 2024-03-08T16:57:10Z
publishDate 2024-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj.art-b717adb2e6d3423a8a23deaf71cea4ff2024-01-05T00:02:59ZengIEEEIEEE Access2169-35362024-01-01121076109110.1109/ACCESS.2023.334668910373019Detection of Obstructive Sleep Apnoea Using Features Extracted From Segmented Time-Series ECG Signals With a One Dimensional Convolutional Neural NetworkSteven Thompson0https://orcid.org/0009-0004-6796-3028Denis Reilly1https://orcid.org/0000-0002-8161-9010Paul Fergus2https://orcid.org/0000-0002-7070-4447Carl Chalmers3https://orcid.org/0000-0003-0822-1150Department of Computer Science, Liverpool John Moores University, Liverpool, U.K.Department of Computer Science, Liverpool John Moores University, Liverpool, U.K.Department of Computer Science, Liverpool John Moores University, Liverpool, U.K.Department of Computer Science, Liverpool John Moores University, Liverpool, U.K.This paper reports on ongoing research, which aims to prove that features of Obstructed Sleep Apnoea (OSA) can be automatically identified from single-lead electrocardiogram (ECG) signals using a One-Dimensional Convolutional Neural Network (1DCNN) model. The 1DCNN is also compared against other machine learning (ML) classifier models, namely Support Vector Machine (SVM) and Random Forest Classifier (RFC). The 1DCNN architecture consists of 4 major parts, a Convolutional Layer, a Flattened Dense Layer, a Max Pooling Layer and a Fully Connected Multilayer Perceptron (MLP), with 1 Hidden Layer and a SoftMax output. The model repeatedly learns how to better extract prominent features from one-dimensional data and map it to the MLP for increased prediction. Training and validation are achieved using pre-processed time-series ECG signals captured from 35 ECG recordings. Using our unique windowing strategy, the data is shaped into 5 datasets of different window sizes. A total of 15 models (5 for each group, 1DCNNs, RFCs, SVMs) were evaluated using various metrics, with each being run over numerous experiments. Results show the 1DCNN-500 model delivered the greatest degree of accuracy and rapidity in comparison to the best producing RFC and SVM classifiers. 1DCNN-500 (Sensitivity 0.9743, Specificity 0.9708, Accuracy 0.9699); RFC-500 (Sensitivity/Recall (0) 0.90 / (1) 0.94, Precision (0) 0.94 / (1) 0.90, Accuracy 0.91); SVM-500 (Sensitivity (0) 0.94 / (1) 0.50, Precision (0) 0.65 / (1) 0.90, Accuracy 0.72). The model presents a novel approach that could provide support mechanisms in clinical practice to promptly diagnose patients suffering from OSA.https://ieeexplore.ieee.org/document/10373019/Apnoea–Hypopnoea index (AHI)electrocardiography (ECG)obstructed sleep Apnoea (OSA)one dimensional convolutional neural network (1DCNN)machine learning (ML)deep learning (DL)
spellingShingle Steven Thompson
Denis Reilly
Paul Fergus
Carl Chalmers
Detection of Obstructive Sleep Apnoea Using Features Extracted From Segmented Time-Series ECG Signals With a One Dimensional Convolutional Neural Network
IEEE Access
Apnoea–Hypopnoea index (AHI)
electrocardiography (ECG)
obstructed sleep Apnoea (OSA)
one dimensional convolutional neural network (1DCNN)
machine learning (ML)
deep learning (DL)
title Detection of Obstructive Sleep Apnoea Using Features Extracted From Segmented Time-Series ECG Signals With a One Dimensional Convolutional Neural Network
title_full Detection of Obstructive Sleep Apnoea Using Features Extracted From Segmented Time-Series ECG Signals With a One Dimensional Convolutional Neural Network
title_fullStr Detection of Obstructive Sleep Apnoea Using Features Extracted From Segmented Time-Series ECG Signals With a One Dimensional Convolutional Neural Network
title_full_unstemmed Detection of Obstructive Sleep Apnoea Using Features Extracted From Segmented Time-Series ECG Signals With a One Dimensional Convolutional Neural Network
title_short Detection of Obstructive Sleep Apnoea Using Features Extracted From Segmented Time-Series ECG Signals With a One Dimensional Convolutional Neural Network
title_sort detection of obstructive sleep apnoea using features extracted from segmented time series ecg signals with a one dimensional convolutional neural network
topic Apnoea–Hypopnoea index (AHI)
electrocardiography (ECG)
obstructed sleep Apnoea (OSA)
one dimensional convolutional neural network (1DCNN)
machine learning (ML)
deep learning (DL)
url https://ieeexplore.ieee.org/document/10373019/
work_keys_str_mv AT steventhompson detectionofobstructivesleepapnoeausingfeaturesextractedfromsegmentedtimeseriesecgsignalswithaonedimensionalconvolutionalneuralnetwork
AT denisreilly detectionofobstructivesleepapnoeausingfeaturesextractedfromsegmentedtimeseriesecgsignalswithaonedimensionalconvolutionalneuralnetwork
AT paulfergus detectionofobstructivesleepapnoeausingfeaturesextractedfromsegmentedtimeseriesecgsignalswithaonedimensionalconvolutionalneuralnetwork
AT carlchalmers detectionofobstructivesleepapnoeausingfeaturesextractedfromsegmentedtimeseriesecgsignalswithaonedimensionalconvolutionalneuralnetwork