Fig 1 - available via license: Creative Commons Attribution-NonCommercial 4.0 International
Content may be subject to copyright.
Source publication
Background:
Heart rate variability (HRV) can reflect the relationship between heart rhythm and sleep structure.
Objective:
In order to study the effect of support vector machine (SVM) on the results of automatic sleep staging and improve the effectiveness of heart rate variability (HRV) as a sleep structure biomarker, thereby realize long term a...
Citations
... In the conventional approach, the foundation of automatic sleep staging lies in the extraction of sleep features and the application of diverse classification algorithms, including support vector machines, random forests, and others. [3][4][5] The researchers selected features ranging from the time domain, frequency domain, to nonlinear characteristics for automatic sleep staging. 6 Multiscale entropy, which exhibited significant differences across distinct sleep stages, was selected as a key feature. ...
Purpose
Sleep structure is crucial in sleep research, characterized by its dynamic nature and temporal progression. Traditional 30-second epochs falter in capturing the intricate subtleties of various micro-sleep states. This paper introduces an innovative artificial neural network model to generate continuous sleep depth value (SDV), utilizing a novel multi-feature fusion approach with EEG data, seamlessly integrating temporal consistency.
Methods
The study involved 50 normal and 100 obstructive sleep apnea–hypopnea syndrome (OSAHS) participants. After segmenting the sleep data into 3-second intervals, a diverse array of 38 feature values were meticulously extracted, including power, spectrum entropy, frequency band duration and so on. The ensemble random forest model calculated the timing fitness value for all the features, from which the top 7 time-correlated features were selected to create detailed sleep sample values ranging from 0 to 1. Subsequently, an artificial neural network (ANN) model was trained to delineate sleep continuity details, unravel concealed patterns, and far surpassed the traditional 5-stage categorization (W, N1, N2, N3, and REM).
Results
The SDV changes from wakeful stage (mean 0.7021, standard deviation 0.2702) to stage N3 (mean 0.0396, standard deviation 0.0969). During the arousal epochs, the SDV increases from the range (0.1 to 0.3) to the range around 0.7, and decreases below 0.3. When in the deep sleep (≤0.1), the probability of arousal of normal individuals is less than 10%, while the average arousal probability of OSA patients is close to 30%.
Conclusion
A sleep continuity model is proposed based on multi-feature fusion, which generates SDV ranging from 0 to 1 (representing deep sleep to wakefulness). It can capture the nuances of the traditional five stages and subtle differences in microstates of sleep, considered as a complement or even an alternative to traditional sleep analysis.
... PSO first initializes a group of random particles then updates the particle state through the optimal particles found in each iteration and the optimal particles found in the entire population, and finally obtains the global optimal solution after the iteration. As a common optimization algorithm, PSO has high efficiency, fast convergence speed, and improved ability for dealing with nonlinear and multi-peak problems [19,20]. In this research, PSO is used to invert the parameters of the inversion optimization model. ...
Atmospheric refraction error is an important factor affecting the positioning accuracy of satellite navigation and radar. In this research, we focused on an inversion correction method based on the mapping function approach to improve the correction accuracy of atmospheric refraction error. First, we analyze the sensitivity of different coefficients on the correction accuracy in the continued fraction function of global mapping function model and conclude that coefficient a in the function is the main factor. Then, the mapping function method is reduced to an inversion correction model with undetermined parameters including coefficient a in the continued fraction function and zenith delay τ, in which the undetermined parameters involve inversion according to observation data by particle swarm optimization so as to improve the correction accuracy of atmospheric refraction error. Finally, application of the inversion method and an example indicate that the inversion correction method can correct the atmospheric refraction error at an impending moment and by using the refraction error under a partial elevation angle.
... The first phase entails deriving attributes from the time waveform, and the second phase involves having a highly trained organizer forecast the sleep phases using the data derived from the waveforms. In terms of sorting, the usual approaches consist of decision trees and arbitrary forests (5), support vector machines (SVM) (6), and neural network (NN)-based methods (7). Alternatively, entropy with numerous scales, auto declining attributes, and linear separator analysis were used by the authors of (8). ...
Background:
We demonstrate an innovative approach of automated sleep recording formed on the electroencephalogram (EEG) with one channel.
Methods:
In this study, double-density dual-tree discrete wavelet transformation (DDDTDWT) was used for decomposing the image, and marginal Fisher analysis (MFA) was used for reducing the dimension. A proposed model on unprocessed EEG models was used on monitored training of 5-group sleep phase forecasting.
Results:
Our network includes a 14-row structure, and a 30-s period was extracted as input in order to be categorized which is followed by second and third period prior to the first 30-s period. Another consecutive period for temporal tissue was added which is not required to a signal preprocess and attribute data derivation phase. Our means of evaluating and improving our approach was to use input from the Sleep Heart Health Study (SHHS), which is a large study field aimed at using research from numerous centers and people and which studies the records of specialist-rated polysomnography (PSG). Performance measures could reach the desired level, which is a precision of 0.87 and a Cohen's kappa of 0.81.
Conclusions:
The use of a large, collaborative study of specialist graders can enhance the likelihood of good globalization. Overall, the novel approach learned by our network showcases the models based on each category.
This paper conducts a study on closed-loop control of engine performance parameters during mode transition process of TBCC engine based on artificial intelligence method. Firstly, a composite modeling method based on stepwise regression analysis and batch normalization-depth neural network is proposed to establish the on-board model during mode transition to estimate the thrust and inlet airflow in real-time. Secondly, based on the hybrid penalty function-particle swarm optimization algorithm, a mode transition control schedule applicable to the closed-loop control of thrust and inlet airflow is developed. Finally, a data processing method based on similarity conversion is proposed to extend the applicable envelope range of the mode transition control system. The transition time is shortened by 33.3 %, and the fluctuations of thrust and inlet airflow are reduced by 1.33 % and 10.77 %, respectively. When the control system is applied to the off-design mode transition process, a satisfactory mode transition performance is also obtained.
