Sleep stage classification using extreme learning machine and particle swarm optimization for healthcare big data

Abstract

Recent developments of portable sensor devices, cloud computing, and machine learning algorithms have led to the emergence of big data analytics in healthcare. The condition of the human body, e.g. the ECG signal, can be monitored regularly by means of a portable sensor device. The use of the machine learning algorithm would then provide an overview of a patient’s current health on a regular basis compared to a medical doctor’s diagnosis that can only be made during a hospital visit. This work aimed to develop an accurate model for classifying sleep stages by features of Heart Rate Variability (HRV) extracted from Electrocardiogram (ECG). The sleep stages classification can be utilized to predict the sleep stages proportion. Where sleep stages proportion information can provide an insight of human sleep quality. The integration of Extreme Learning Machine (ELM) and Particle Swarm Optimization (PSO) was utilized for selecting features and determining the number of hidden nodes. The results were compared to Support Vector Machine (SVM) and ELM methods which are lower than the integration of ELM with PSO. The results of accuracy tests for the combined ELM and PSO were 62.66%, 71.52%, 76.77%, and 82.1% respectively for 6, 4, 3, and 2 classes. To sum up, the classification accuracy can be improved by deploying PSO algorithm for feature selection.

Entropy Analysis of Heart Rate Variability in Different Sleep Stages

“…The third reason may be that we did not try to optimize the training and input features. By comparison, Surantha et al [59] selected different features for every sample for staging that achieved an accuracy of 71.52%. This is, however, difficult to implement in practical applications.…”

An Enhanced Posture Prediction-Bayesian Network Algorithm for Sleep Posture Recognition in Wireless Body Area Networks

“…The comparative study shown in Table 1 considers various datasets by which the performance varies in terms of accuracy. Hence, the comparison is mainly focused on the postures considered in the initial works considering the off-bed, sitting, lying center, lying left, and lying right, whereby the prone posture is not considered in deep neural networks and K -nearest neighborhood algorithm [ 11 ]. The next important comparison parameter is the number of sensors used.…”

“…A low-cost pressure array is embedded with conductive fabric and conductive wires, where sensors are deployed with bedspreads configured as 32 rows and 32 columns with a total of 1024 nodes. The paper considers Arduino nano for collecting data using a 10-bit analog-digital converter (ADC) [ 11 ]. The data acquisition was done considering six sleeping postures from five participants, with the convolutional neural networks implemented on a PC.…”

“…This method is an efficient way to extract the objects for object detection that removes the offset of weights, position, and size of different objects. The filters are applied in horizontal and vertical directions by predefining the size of the cells and the number of cells in one block [ 11 ]. This work focuses on the weight and space normalization of images.…”

Three Steps to Improve Jellyfish Search Optimiser

“…The JS algorithm belongs to swarm optimisation methods which are inspired by biosystems from nature. Swarm algorithms are very popular in practical (real-world) applications in various areas of research, industry and healthcare services [7,15].…”