Study of background complexity based on composite multiscale entropy of EEG signals

The cardiovascular system (CVS) regulation can be studied from a central viewpoint, through heart rate variability (HRV) data, and from a peripheral viewpoint, through laser Doppler flowmetry (LDF) signals. Both the central and peripheral CVSs are regulated by several interacting mechanisms, each having its own temporal scale. The central CVS has been the subject of many multiscale studies. By contrast, these studies at the level of the peripheral CVS are very recent. Among the multiscale studies performed on the central CVS data, multiscale entropy has been proven to give interesting physiological information for diagnostic purposes. However, no multiscale entropy analysis has been performed on LDF signals. The authors' goal is therefore to propose a first multiscale entropy study of LDF data recorded in healthy subjects. The LDF signals recorded in the forearm of seven healthy subjects are processed. Their period sampling is T=50 ms, and coarse-graining scales from T to 23T are studied. Also, for validation, the algorithm is first tested on synthetic signals of known theoretical multiscale entropy. The results reveal nonmonotonic evolution of the multiscale entropy of LDF signals, with a maximum at small scales around 7T and a minimum at longer scales around 18T, singling out in this way two distinctive scales where the LDF signals undergo specific changes from high to low complexity. This also marks a strong contrast with the HRV signals that usually display a monotonic increase in the evolution of the multiscale entropy. Multiscale entropy of LDF signals in healthy subjects shows variation with scales. Moreover, as the variation pattern observed appears similar for all the tested signals, multiscale entropy could potentially be a useful stationary signature for LDF signals, which otherwise are probe-position and subject dependent. Further work could now be conducted to evaluate possible diagnostic purposes of the multiscale entropy of LDF signals.

Objective: The purpose of this study was to evaluate whether the complexity of the corneal pulse (CP) signal can be used to differentiate patients with primary open-angle glaucoma (POAG) from healthy subjects. Approach: The study sample consisted of 28 patients with POAG and a control, age-matched group of 30 subjects. After standard ophthalmic examination, the CP signal from a randomly selected eye of each participant was measured using non-contact ultrasonic micro-displacement measurement technology. After pre-processing, the complexity of the CP signal was estimated using refined composite multiscale fuzzy entropy (RCMFE) up to scale factor 50. The average RCMFE values were computed from three repeated measurements of the CP signals for each participant and each scale factor. Main results: The complexity of the CP signal in glaucomatous eyes was higher than that observed in healthy ones. Also, RCMFE of the CP signal was found to differentiate (statistically significantly) between the two groups for scales in the range from 26 to 43. For these scales, the one for which the lowest p-value (t-test, p = 0.017) was obtained when comparing RCMFE between the two groups was selected as the optimal scale. Next, a receiver operating characteristic analysis for the optimal scale showed that the proposed approach of calculating the multiscale entropy of the CP signal has some potential to discriminate between patients with POAG and healthy controls (sensitivity, specificity and accuracy of 0.643, 0.700 and 0.672, respectively). Significance: In conclusion, RCMFE, as a complexity measure, may be considered an auxiliary indicator to support glaucoma diagnostics.

Subliminal stimulation can trigger emotional responses and this paper analyzes their relationship based on EEG signals. According to the nonlinear characteristics of EEG signals and based on the multi-scale sample entropy, we propose a method to recognize unconscious emotion. We extract the multi-scale sample entropy of EEG signals when angry or happy face pictures are subliminally presented as eigenvalues. Then, we compute the distribution and regularity of eigenvalues. Experimental results demonstrate that the multi-scale sample entropy can be used to distinguish different emotions when different emotional pictures are subliminally presented, namely, the multi-scale sample entropy is larger when angry face pictures are subliminally presented than happy face pictures. Furthermore, the p-values of extracted features are calculated by the method of Kolmogorov-Smirnov test (KS test). The result shows that the EEG signals have significant difference $(\mathrm{p}\lt 0.05)$ for the two different emotional states and the method is effective for unconscious emotion recognition.

This paper analyzes the underlying complexity and non-linearity of electroencephalogram (EEG) signals by computing a novel multi-scale entropy measure for the classification of seizure, seizure-free and normal EEG signals. The quality factor (Q) based multi-scale entropy measure is proposed to compute the entropy of the EEG signal in different frequency-bands of interest. The Q -based entropy (QEn) is computed by decomposing the signal with the tunable-Q wavelet transform (TQWT) into the number of sub-bands and estimating K-nearest neighbor (K-NN) entropies from various sub-bands cumulatively. The optimal selection of Q and the redundancy parameter (R) of TQWT showed better robustness for entropy computation in the presence of high- and low-frequency components. The extracted features are fed to the support vector machine (SVM) classifier with the wrapper-based feature selection method. The proposed method has achieved accuracy of 100% in classifying normal (eyes-open and eyes-closed) and seizure EEG signals, 99.5% in classifying seizure-free EEG signals (from the hippocampal formation of the opposite hemisphere of the brain) from seizure EEG signals and 98% in classifying seizure-free EEG signals (from the epileptogenic zone) from seizure EEG signals, respectively, using the SVM classifier. We have also achieved classification accuracies of 99% and 98.6% in classifying seizure versus non-seizure EEG signals and the individual three classes, namely normal, seizure-free and seizure EEG signals, respectively. The performance measure of the proposed multi-scale entropy has been found to be comparable with the existing state of the art epileptic EEG signals classification methods studied using the same database.

Alzheimer disease (AD) is the most prevalent neurodegenerative disorder and a type of dementia. About 80% of dementia in older adults is due to AD. According to multiple research articles, AD is associated with several changes in EEG signals, such as slow rhythms, reduction in complexity and functional associations, and disordered functional communication between different brain areas. This research focuses on the entropy parameter. In this study, the keywords "Entropy," "EEG," and "Alzheimer" were used. In the initial search, 102 articles were found. In the first stage, after investigating the Abstracts of the articles, the number of them was reduced to 62, and upon further review of the remaining articles, the number of articles was reduced to 18. Some papers have used more than one entropy of EEG signals to compare, and some used more than one database. So, 25 entropy measures were considered in this meta-analysis. We used the Standardized Mean Difference (SMD) to find the effect size and compare the effects of AD on the entropy of the EEG signal in healthy people. Funnel plots were used to investigate the bias of meta-analysis. According to the articles, entropy seems to be a good benchmark for comparing the EEG signals between healthy people and AD people. It can be concluded that AD can significantly affect EEG signals and reduce the entropy of EEG signals. Our primary question addressed in this study is "Can Alzheimer's Disease significantly affect EEG signals or not?" This paper is the first Meta-Analysis study that reveals the effects of Alzheimer's Disease on EEG signals and the caused reduction in the complexity of the EEG signal. According to the articles, results and funnel plots of this Meta-Analysis, entropy seems to be a good benchmark for comparing the EEG signals in healthy people and people who have Alzheimer's Disease. Alzheimer's Disease is one of the most prevalent neurodegenerative disorder which can affect EEG signals. This study is the first Meta-Analysis in this regard and the results confirm that Alzheimer's Disease reduces the complexity of the EEG signals. We used 25 entropy measures applied in 18 articles. The materials in this Meta-Analysis are 1-SMD for finding the effect size and 2- Funnel plot for investigating the bias of Meta-Analysis.

As alternative entropy estimators, multiscale entropy (MSE) and permutation entropy (PE) are utilized for quantification of the brain function and its signal variability. In this context, their applications are primarily focused on two specific domains: (1) the effect of brain pathology on its function (2) the study of altered states of consciousness. As a result, there is a paucity of research on applicability of these measures in more naturalistic scenarios. In addition, the utility of these measures for quantification of the brain function and with respect to its signal entropy is not well studied. These shortcomings limit the interpretability of the measures when used for quantification of the brain signal entropy. The present study addresses these limitations by comparing MSE and PE with entropy of human subjects’ EEG recordings, who watched short movie clips with negative, neutral, and positive content. The contribution of the present study is threefold. First, it identifies a significant anti-correlation between MSE and entropy. In this regard, it also verifies that such an anti-correlation is stronger in the case of negative rather than positive or neutral affects. Second, it finds that MSE significantly differentiates between these three affective states. Third, it observes that the use of PE does not warrant such significant differences. These results highlight the level of association between brain’s entropy in response to affective stimuli on the one hand and its quantification in terms of MSE and PE on the other hand. This, in turn, allows for more informed conclusions on the utility of MSE and PE for the study and analysis of the brain signal variability in naturalistic scenarios.

Multiscale entropy (MSE) of physiological signals may reflect cardiovascular health in diabetes. The classic MSE (cMSE) algorithm requires more than 750 signals for the calculations. The modified short-time MSE (sMSE) may have inconsistent outcomes compared with the cMSE at large time scales and in a disease status. Therefore, we compared the cMSE of 1500 (cMSE1500) consecutive and 1000 photoplethysmographic (PPG) pulse amplitudes with the sMSE of 500 PPG (sMSE500) pulse amplitudes of bilateral fingertips among middle- to old-aged individuals with or without type 2 diabetes. We discovered that cMSE1500 had the smallest value across scale factors 1–10, followed by cMSE1000, and then sMSE500 in both hands. The cMSE1500, cMSE1000 and sMSE500 did not differ at each scale factor in both hands of persons without diabetes and in the dominant hand of those with diabetes. In contrast, the sMSE500 differed at all scales 1–10 in the non-dominant hand with diabetes. In conclusion, autonomic dysfunction, prevalent in the non-dominant hand which had a low local physical activity in the person with diabetes, might be imprecisely evaluated by the sMSE; therefore, using more PPG signal numbers for the cMSE is preferred in such a situation.

When the driver is driving the vehicle, there are three driving behaviors. They are turning left, turning right, and going straight. In order to find out the difference of different driving behaviors on electroencephalogram (EEG) signals, an experimental method was designed to collect the EEG signals from different driving behaviors of drivers. The EEG signals were analyzed and studied by nonlinear dynamics. Firstly, the EEG signals were sub-band decomposed by wavelet transform to extract the alpha wave and the artifacts of the signals were removed. Then, the processed EEG signals from different driving behaviors were calculated to get the complexity and approximate entropy by the Lempel-Ziv complexity algorithm and the approximate entropy algorithm in nonlinear dynamics. The calculation results showed that the complexity value and approximate entropy of the EEG signal from turning left were the largest, and the complexity and approximate entropy of the EEG signal from going straight were the smallest. The result indicated that among the three driving behaviors, the EEG signals of drivers from turning left was relatively more complex and the most random. However, the change rate of the EEG signal complexity and approximate entropy in the three driving behaviors was not large, and the difference was not significant.

Multiscale base-scale entropy is introduced in this paper.We use it to analyze heart rate variability series.The results show that multiscale base-scale entropy can identify patterns generated from healthy and pathologic states, and can distinguish daytime and nighttime heartbeat time series. We also calculate the multiscale base-scale entropy of surrogate signal (phase randomized data), compare it with the entropy of atrial fibrillation signal, and find that the tends of two entropys are similar to each other, which indicates that atrial fibrillation reflects the linear characteristics of physiological signals. Multiscale base-scale entropy method has potential applications to studying a wide variety of other physiologic and physical time series data.

The neural mechanisms that underpin cognitive impairments in patients with schizophrenia remain unclear. Previous studies have typically treated patients as a homogeneous group, despite the existence of distinct symptom presentations between deficit and non-deficit subtypes. This approach has been found to be inadequate, necessitating separate investigation. This study was conducted at Daizhuang Hospital in Jining City, China, from January 2022 to October 2023. The study sample comprised 30 healthy controls, 19 patients with deficit schizophrenia, and 19 patients with non-deficit schizophrenia, all aged between 18 and 45 years. Cognitive abilities were evaluated using a change detection task. The NeuroScan EEG/ERP System, comprising 64 channels and utilising standard 10-20 electrode placements, was employed to record EEG signals. The multiscale entropy and sample entropy of the EEG signals were calculated. The healthy controls demonstrated superior task performance compared to both the non-deficit (p < 0.001) and deficit groups(p < 0.001). Significant differences in multiscale entropy between the three groups were observed at multiple electrode sites. In the task state, there are significant differences in the sample entropy of the β frequency band among the three groups of subjects. Under simple conditions of difficulty, the performance of the healthy controls exhibited a positive correlation with alpha band sample entropy(r = 0.372) and a negative correlation with beta band sample entropy (r = -0.411). Deficit patients demonstrated positive correlations with alpha band sample entropy (r = 0.370), whereas non-deficit patients exhibited negative correlations with both alpha and beta band sample entropy (r = -0.451, r = -0.362). Under difficult conditions of difficulty, the performance of healthy controls demonstrated a positive correlation with beta band sample entropy (r = 0.486). Deficit patients exhibited a positive correlation with alpha band sample entropy (r = 0.351), while non-deficit patients demonstrated a negative correlation with beta band sample entropy (r = -0.331). The results of this study indicate that cognitive impairment in specific subtypes of schizophrenia may have distinct physiological underpinnings, underscoring the need for further investigation.

This study aims to measure Depth of Anesthesia(DoA) using non linear Electroencephalogram (EEG) signal analysis and Wavelet Analysis during anesthesia. Two features Approximate Entropy(ApEn) and Wavelet Entropy(WE) of the EEG signals were extracted as a measure of DoA from the EEG signals during four phases of general anesthesia called awake, induction, maintenance and recovery. In order to find out wavelet entropy, EEG signals during anesthesia were decomposed into its constituent frequency bands, then WE is calculated from the approximation and detail coefficients. Approximate Entropy is calculated from the respective algorithm. Finally these two extracted DoA measures were compared with BIS index, which is a commercially available DoA monitor.

It is well established that the e4 allele of the APOE gene is associated with impaired brain functionality and cognitive decline in humans at elder age. However, it is controversial whether and how the APOE e4 allele is associated with superior brain function among young healthy individuals, thus indicates a case of antagonistic pleiotropy of APOE e4 allele. Signal complexity is a critical aspect of brain activity that has been associated with brain function. In this study, the multiscale entropy (MSE) of resting-state EEG signals among a sample of young healthy adults (N = 260) as an indicator of brain signal complexity was investigated. It was of interest whether MSE differs across APOE genotype groups while age and education level were controlled for and whether the APOE genotype effect on MSE interacts with MSE time scale, as well as EEG recording condition. Results of linear mixed models indicate overall larger MSE in APOE e4 carriers. This genotype-dependent difference is larger at high as compared with low time scales. The interaction effect between APOE genotype and recording condition indicates increased between-state MSE change in young healthy APOE e4 carriers as compared with non-carriers. Because higher complexity is commonly taken to be associated with better cognitive functioning, the present results complement previous findings and therefore point to a pleiotropic spectrum of the APOE gene polymorphism.

Entropy is an approach for computing uncertainty. Over recent decades, entropy has been developed to measure complexity in biological signals. With the development of entropy, the diversity of its application has increased. In civil engineering, entropy has been combined with structural health monitoring (SHM) for damage detection. Moreover, as algorithms for entropy have continued to be developed, multiscale entropy (MSE), composite multiscale entropy (CMSE), and composite multiscale cross-sample entropy (CMSCE) have been proposed in succession. The aim of this study was to optimize CMSCE to enhance SHM performance. To reduce the influence of ambient noise, ensemble empirical mode decomposition (EEMD) can be used to filter structural dynamic signals. Therefore, the first mode of structure was extracted by using EEMD for entropy analysis and evaluation of damage assessment performance. A numerical simulation of a seven-story steel structure was run to verify the efficacy of EEMD and calculate damage indices for detection of damaged locations. Through this simulation, signals with and without EEMD were compared. As the result, it can be observed that the performance of damage identification was improved in low floors by CMSCE with EEMD.

Automated detection of epileptic seizures using multiscale and refined composite multiscale dispersion entropy

Target-directed elbow movements are essential in daily life; however, how different task demands affect motor control is seldom reported. In this study, the relationship between task demands and the complexity of kinematics and electromyographic (EMG) signals on healthy young individuals was investigated. Tracking tasks with four levels of task demands were designed, and participants were instructed to track the target trajectories by extending or flexing their elbow joint. The actual trajectories and EMG signals from the biceps and triceps were recorded simultaneously. Multiscale fuzzy entropy was utilized to analyze the complexity of actual trajectories and EMG signals over multiple time scales. Results showed that the complexity of actual trajectories and EMG signals increased when task demands increased. As the time scale increased, there was a monotonic rise in the complexity of actual trajectories, while the complexity of EMG signals rose first, and then fell. Noise abatement may account for the decreasing entropy of EMG signals at larger time scales. This study confirmed the uniqueness of multiscale entropy, which may be useful in the analysis of electrophysiological signals.