Ensemble of Deep Learning Models for Sleep Apnea Detection: An Experimental Study

Debadyuti Mukherjee,Ram Sarkar,Friedhelm Schwenker,Koustav Dhar

doi:10.3390/s21165425

Debadyuti Mukherjee, Ram Sarkar + Show 2 more

Open Access

https://doi.org/10.3390/s21165425

Copy DOI

Abstract

Sleep Apnea is a breathing disorder occurring during sleep. Older people suffer most from this disease. In-time diagnosis of apnea is needed which can be observed by the application of a proper health monitoring system. In this work, we focus on Obstructive Sleep Apnea (OSA) detection from the Electrocardiogram (ECG) signals obtained through the body sensors. Our work mainly consists of an experimental study of different ensemble techniques applied on three deep learning models—two Convolutional Neural Network (CNN) based models, and a combination of CNN and Long Short-Term Memory (LSTM) models, which were previously proposed in the OSA detection domain. We have chosen four ensemble techniques—majority voting, sum rule and Choquet integral based fuzzy fusion and trainable ensemble using Multi-Layer Perceptron (MLP) for our case study. All the experiments are conducted on the benchmark PhysioNet Apnea-ECG Database. Finally, we have achieved highest OSA detection accuracy of 85.58% using the MLP based ensemble approach. Our best result is also able to surpass many of state-of-the-art methods.

Highlights

In this modern era, the role of health monitoring systems is increasing in our daily life
We used five classification measures—(i) accuracy, (ii) precision, (iii) recall, (iv) F1-score, (v) specificity to evaluate the performance of the base models and their ensemble
We defined the five chosen classification metrics based on the terms True Positive (TP), True Negative (TN), False Positive (FP) and False Negative (FN)

Summary

Introduction

The role of health monitoring systems is increasing in our daily life. Older people are the most benefited ones from the merits of monitoring their health. Under any healthcare system attached to various body parts, can sense and record the required features of the human body. These kinds of sensors can be placed in any smart-watch or smartphone. The introduction of the Internet of Things (IoT) in the healthcare domain has further upgraded the facilities [1]. Health-based alarms, personal smart medical recommendations, etc. Have decreased the life-risks caused by sudden health problems Health-based alarms, personal smart medical recommendations, etc. have decreased the life-risks caused by sudden health problems

Methods

Results

Conclusion