Sleep staging algorithm based on multichannel data adding and multifeature screening

Huang, Wu; Guo, Bing; Shen, Yan; Tang, Xiangdong; Zhang, Tao; Li, Dan; Jiang, Zhonghui

doi:10.1016/j.cmpb.2019.105253

Cited by 33 publications

(13 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…After data collection for the two devices was completed, the output data were transformed into reports with the respective analysis software and output, as shown in Figure 6 A,B [ 15 , 16 , 34 ]. Large peaks in motion tracking ( Figure 6 A) generally occurred at the same times for both ActiWatch2 and our device.…”

Section: Resultsmentioning

confidence: 99%

“…For advanced analyses, the data were analyzed with the output data of the environmental light sensor as a moderator, as shown in Figure 3. The data processing and analysis were performed using MATLAB ® (R2007a, MathWorks Inc., Natick, MA, USA), and the results were visualized as chart-style reports [34]. The addition of the ambient light data allowed us to accurately determine sleep periods from the motion detection data.…”

Section: Core Functionmentioning

confidence: 99%

See 1 more Smart Citation

Design and Implementation of a Multifunction Wearable Device to Monitor Sleep Physiological Signals

et al. 2020

View full text Add to dashboard Cite

We present a wearable device built on an Adafruit Circuit Playground Express (CPE) board and integrated with a photoplethysmographic (PPG) optical sensor for heart rate monitoring and multiple embedded sensors for medical applications—in particular, sleep physiological signal monitoring. Our device is portable and lightweight. Due to the microcontroller unit (MCU)-based architecture of the proposed device, it is scalable and flexible. Thus, with the addition of different plug-and-play sensors, it can be used in many applications in different fields. The innovation introduced in this study is that with additional sensors, we can determine whether there are intermediary variables that can be modified to improve our sleep monitoring algorithm. Additionally, although the proposed device has a relatively low cost, it achieves substantially improved performance compared to the commercially available Philips ActiWatch2 wearable device, which has been approved by the Food and Drug Administration (FDA). To assess the reliability of our device, we compared physiological sleep signals recorded simultaneously from volunteers using both our device and ActiWatch2. Motion and light detection data from our device were shown to be correlated to data simultaneously collected using the ActiWatch2, with correlation coefficients of 0.78 and 0.89, respectively. For 7 days of continuous data collection, there was only one instance of a false positive, in which our device detected a sleep interval, while the ActiWatch2 did not. The most important aspect of our research is the use of an open architecture. At the hardware level, general purpose input/output (GPIO), serial peripheral interface (SPI), integrated circuit (I2C), and universal asynchronous receiver-transmitter (UART) standards were used. At the software level, an object-oriented programming methodology was used to develop the system. Because the use of plug-and-play sensors is associated with the risk of adverse outcomes, such as system instability, this study heavily relied on object-oriented programming. Object-oriented programming improves system stability when hardware components are replaced or upgraded, allowing us to change the original system components at a low cost. Therefore, our device is easily scalable and has low commercialization costs. The proposed wearable device can facilitate the long-term tracking of physiological signals in sleep monitoring and related research. The open architecture of our device facilitates collaboration and allows other researchers to adapt our device for use in their own research, which is the main characteristic and contribution of this study.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Core Functionmentioning

confidence: 99%

Design and Implementation of a Multifunction Wearable Device to Monitor Sleep Physiological Signals

et al. 2020

View full text Add to dashboard Cite

show abstract

“…For Sleep-EDF dataset, we examined the number of important features to be selected with 50 increments in between 500 and 2500, and the most appropriate feature subset were determined for each number of classes. More details about the feature selection in this study are found in [27] which describes basically the same feature selection method.…”

Section: Feature Selection and Optimizationmentioning

confidence: 99%

“…The results of applying our method against the latest, extended version of the Sleep-EDF database, show that in contrast to the first version of the database which consisted of 61 recordings (version 1), the latest version consists of 197 recordings (version 2, released in 2018). It has been studied in many recent papers (e.g., [27,28], however, because of its large size, it is rarely studied as a whole (many papers which classified it are using only a small subset of it). Therefore, it is hard to compare the performance on it under the same or similar conditions.…”

Section: Classification Of Sleep-edf Datasetmentioning

confidence: 99%

Classification of Brainwaves for Sleep Stages by High-Dimensional FFT Features from EEG Signals

et al. 2020

View full text Add to dashboard Cite

Manual classification of sleep stage is a time-consuming but necessary step in the diagnosis and treatment of sleep disorders, and its automation has been an area of active study. The previous works have shown that low dimensional fast Fourier transform (FFT) features and many machine learning algorithms have been applied. In this paper, we demonstrate utilization of features extracted from EEG signals via FFT to improve the performance of automated sleep stage classification through machine learning methods. Unlike previous works using FFT, we incorporated thousands of FFT features in order to classify the sleep stages into 2–6 classes. Using the expanded version of Sleep-EDF dataset with 61 recordings, our method outperformed other state-of-the art methods. This result indicates that high dimensional FFT features in combination with a simple feature selection is effective for the improvement of automated sleep stage classification.

show abstract

“…Among them, Yücelbaş et al (2015) went for the purification of EEG signals for a healthy evaluation due to the noises of EEG, EOG, EMG signals during recording and measured the staging success between the pure EEG signal and the impure EEG signal using ANN (Yücelbaş et al, 2015). Huang et al (2020) have suggested a signal pre-processing and feature scanning method. They classified the features which are obtained with SVM method (Huang et al, 2020).…”

Section: Introductionmentioning

confidence: 99%

PSG Kayıt Sinyalleri Kullanılarak Uyku Evrelerinin Sınıflandırılması

Koca

Özşen²,

Göğüş

et al. 2020

European Journal of Science and Technology

View full text Add to dashboard Cite

Automatic sleep staging is aimed within the scope of this paper. Sleep staging is a study by a sleep specialist. Since this process takes quite a long time and sleep is a method based on the knowledge and experience, it is inevitable for each person to show different results. For this, an automatic sleep staging method has been introduced. In the study, EEG (Electroencephalogram), EOG (Electrooculogram), EMG (Electromyogram) data recorded by PSG (Polysomnography) device for seven patients in Necmettin Erbakan University sleep laboratory were used. 81 different features were taken from the data in time and frequency environment. Also, PCA (Principal component analysis) and SFS (Sequential forward selection) feature selection methods were used. The classification success of the sleep phases in different machine learning methods was measured by using the received features. Linear D. (Linear Discriminant Analysis), Cubic SVM (Support vector machine), Weighted kNN (k nearest neighbor), Bagged Trees, ANN (Artificial neural network) were used as classifiers. System success was achieved with a 5 fold cross-validation method. Accuracy rates obtained were respectively 55.6%, 65.8%, 67%, 72.1%, and 69.1%.

show abstract

Sleep staging algorithm based on multichannel data adding and multifeature screening

Cited by 33 publications

References 37 publications

Design and Implementation of a Multifunction Wearable Device to Monitor Sleep Physiological Signals

Design and Implementation of a Multifunction Wearable Device to Monitor Sleep Physiological Signals

Classification of Brainwaves for Sleep Stages by High-Dimensional FFT Features from EEG Signals

PSG Kayıt Sinyalleri Kullanılarak Uyku Evrelerinin Sınıflandırılması

Contact Info

Product

Resources

About