Bi-directional Long Short-Term Memory using Quantized data of Deep Belief Networks for Sleep Stage Classification

Yulita, Intan Nurma; Fanany, Mohamad Ivan; Arymuthy, Aniati Murni

doi:10.1016/j.procs.2017.10.042

Cited by 46 publications

(25 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Firstly, the automatic sleep staging methods based on EEG signals have realized considerable performance since there is a direct link between EEG and electrophysiological activities of the brain [4]. The accuracy of 5-class sleep staging using single-lead EEG for healthy people can exceed 90% [5], [6]; 5-class sleep staging based on EEG, EMG, and EOG can achieve the accuracy of 92% or even higher for healthy people [7], [8], and 86% accuracy for patients with sleep disorders [9]. Secondly, the sleep staging methods based on cardiopulmonary coupling signals that mainly contain ECG and respiratory signals have attracted more and more attention.…”

Section: Introductionmentioning

confidence: 99%

“…So LSTM can utilize the temporal correlation of time series and avoid the problem of long-term dependence [22]. Yulita et al used Bidirectional LSTM for sleep staging using EEG, EOG, and EMG signals, achieving an accuracy of 86% for patients with sleep disorders [9], [23]. Radha et al used the LSTM model to study sleep staging for healthy people and also yielded good results [24].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Multi-Class Automatic Sleep Staging Method Based on Long Short-Term Memory Network Using Single-Lead Electrocardiogram Signals

et al. 2019

View full text Add to dashboard Cite

To overcome the disadvantage of clinical manual sleep staging, a convenient, economical, and efficient multi-class automatic sleep staging method is proposed based on long short-term memory network (LSTM) using single-lead electrocardiogram signals. From electrocardiogram signals, heart rate variability and respiratory signals were calculated, and, then, totally 25 features were extracted. Four different classifiers, including the two-class classifier to distinguish between wake and sleep, the three-class classifier to distinguish wake, non-rapid eye movement sleep, and rapid eye movement, the four-class classifier to distinguish wake, light sleep, slow wave sleep, and rapid eye movement, and the five-class classifier to distinguish wake, sleep stage N1, sleep stage N2, sleep stage N3, and rapid eye movement, were constructed using the LSTM. The single-lead electrocardiogram data from 238 patients with full sleep stages during sleep were used for the training set and the data from other 60 patients were regarded as a validation set. The rest of 75 patients have left aside for testing set. The accuracy of two-class, three-class, four-class, and five-class sleep staging was 89.84%, 84.07%, 77.76%, and 71.16% and the Cohen's kappa statistic k was 0.52, 0.58, 0.55, and 0.52, respectively, which realized the moderate agreement with clinical analysis. When expanding the dataset to extra 1068 patients with missing sleep stages, the accuracy has no obvious reduction but the Cohen's kappa statistic k dropped to 0.51, 0.52, 0.48, and 0.43, respectively. The proposed method, in this paper, is promising for low-cost, efficient, and convenient sleep staging in home care monitoring. INDEX TERMS Electrocardiogram, heart rate variability, long short-term memory, sleep staging.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A Multi-Class Automatic Sleep Staging Method Based on Long Short-Term Memory Network Using Single-Lead Electrocardiogram Signals

et al. 2019

View full text Add to dashboard Cite

show abstract

“…In 2016, they make a comparison between RNAs and SVMs to classify flaws in roller production, obtaining as a result that SMVs provide better results. Similarly, [16] and use SVMs to classify skin diseases [15] and present a deep belief network (DBN) to solve a classification problem about sleep quality [22][23].…”

Section: Related Researchmentioning

confidence: 99%

A Genetic Programming Approach to Binary Classification Problem

Santoso¹,

Singh²,

Rajest³

et al. 2018

EAI Endorsed Transactions on Energy Web

View full text Add to dashboard Cite

The Binary classification is the most challenging problem in machine learning. One of the most promising technique to solve this problem is by implementing genetic programming (GP). GP is one of Evolutionary Algorithm (EA) that used to solve problems that humans do not know how to solve it directly. The objectives of this research is to demonstrate the use of genetic programming in this type of problems; that is, other types of techniques are typically used, e.g., regression, artificial neural networks. Genetic programming presents an advantage compared to those techniques, which is that it does not need an a priori definition of its structure. The algorithm evolves automatically until finding a model that best fits a set of training data. Feature engineering was considered to improve the accuracy. In this research, feature transformation and feature creation were implemented. Thus, genetic programming can be considered as an alternative option for the development of intelligent systems mainly in the pattern recognition field.

show abstract

“…Correspondingly, intelligent diagnosis methods have gradually become mainstream. On account of the excellent data processing capabilities, many methods based on artificial intelligence have gradually been employed in the territory of mechanical fault diagnosis, such as convolutional neural networks (CNN) [ 7 , 8 , 9 ], autoencoder [ 9 , 10 ], deep belief networks [ 11 , 12 ], and recurrent neural networks [ 13 , 14 ].…”

Section: Introductionmentioning

confidence: 99%

Intelligent Fault Diagnosis of Hydraulic Piston Pump Based on Wavelet Analysis and Improved AlexNet

Zhu

Wang

et al. 2021

Sensors

View full text Add to dashboard Cite

Hydraulic piston pump is the heart of hydraulic transmission system. On account of the limitations of traditional fault diagnosis in the dependence on expert experience knowledge and the extraction of fault features, it is of great meaning to explore the intelligent diagnosis methods of hydraulic piston pump. Motivated by deep learning theory, a novel intelligent fault diagnosis method for hydraulic piston pump is proposed via combining wavelet analysis with improved convolutional neural network (CNN). Compared with the classic AlexNet, the proposed method decreases the number of parameters and computational complexity by means of modifying the structure of network. The constructed model fully integrates the ability of wavelet analysis in feature extraction and the ability of CNN in deep learning. The proposed method is employed to extract the fault features from the measured vibration signals of the piston pump and realize the fault classification. The fault data are mainly from five different health states: central spring failure, sliding slipper wear, swash plate wear, loose slipper, and normal state, respectively. The results show that the proposed method can extract the characteristics of the vibration signals of the piston pump in multiple states, and effectively realize intelligent fault recognition. To further demonstrate the recognition property of the proposed model, different CNN models are used for comparisons, involving standard LeNet-5, improved 2D LeNet-5, and standard AlexNet. Compared with the models for contrastive analysis, the proposed method has the highest recognition accuracy, and the proposed model is more robust.

show abstract

Bi-directional Long Short-Term Memory using Quantized data of Deep Belief Networks for Sleep Stage Classification

Cited by 46 publications

References 7 publications

A Multi-Class Automatic Sleep Staging Method Based on Long Short-Term Memory Network Using Single-Lead Electrocardiogram Signals

A Multi-Class Automatic Sleep Staging Method Based on Long Short-Term Memory Network Using Single-Lead Electrocardiogram Signals

A Genetic Programming Approach to Binary Classification Problem

Intelligent Fault Diagnosis of Hydraulic Piston Pump Based on Wavelet Analysis and Improved AlexNet

Contact Info

Product

Resources

About