Discrimination of Severity of Alzheimer’s Disease with Multiscale Entropy Analysis of EEG Dynamics

Abstract

Multiscale entropy (MSE) was used to analyze electroencephalography (EEG) signals to differentiate patients with Alzheimer’s disease (AD) from healthy subjects. It was found that the MSE values of the EEG signals from the healthy subjects are higher than those of the AD ones at small time scale factors in the MSE algorithm, while lower than those of the AD patients at large time scale factors. Based on the finding, we applied the linear discriminant analysis (LDA) to optimize the differentiating performance by comparing the resulting weighted sum of the MSE values under some specific time scales of each subject. The EEG data from 15 healthy subjects, 69 patients with mild AD, and 15 patients with moderate to severe AD were recorded. As a result, the weighted sum values are significantly higher for the healthy than the patients with moderate to severe AD groups. The optimal testing accuracy under five specific scales is 100% based on the EEG signals acquired from the T4 electrode. The resulting weighted sum value for the mild AD group is in the middle of those for the healthy and the moderate to severe AD groups. Therefore, the MSE-based weighted sum value can potentially be an index of severity of Alzheimer’s disease.

Brain Dynamics Altered by Photic Stimulation in Patients with Alzheimer’s Disease and Mild Cognitive Impairment

“…Fifth, this study used sample entropy [ 12 ] as the entropic metric of multiscale entropy analysis [ 10 , 34 ] to quantify EEG complexity in AD, MCI patients, and elderly controls, as widely adopted by clinical studies investigating MSE-based nonlinear analysis of EEG signals in AD [ 14 , 33 , 34 , 35 , 36 , 37 , 39 , 40 , 41 ]. This allows straightforward comparisons of the MSE results and for discussing their interpretations related to AD.…”

“…It has been posited that this is the result of alpha slowing due to thalamocortical dysrhythmia [ 19 , 20 , 21 , 22 , 23 , 24 ]. In order to investigate changes in nonlinear intrinsic brain complexity, MSE-based analyses have been applied to resting-state EEG and MEG signals of individuals with different pathological conditions [ 9 , 25 ], such as psychiatric disorders [ 26 , 27 ], pain conditions [ 28 , 29 , 30 , 31 ], and neurological diseases [ 32 , 33 , 34 , 35 , 36 , 37 , 38 ]. In particular, MSE-based resting-state EEG studies in AD and dementia patients [ 33 , 34 , 35 , 36 , 37 , 39 , 40 , 41 ] have reported that at small time scales, entropy values were lower in MCI and AD patients than in healthy controls; at large time scales, entropy values tended to be higher in AD patients than in healthy controls.…”

Investigating the Interaction Between EEG and HRV Protocol for Study of Three-Circle Post Standing Qigong

“…However, experimental evidence to support this hypothesis is limited. In this study, we are going to evaluate whether the widely utilized complexity-based metrics-MSE is sensitive and informative metric for characterizing the impact of qigong on complexity in cardiovascular and nervous systems [54][55][56] .…”

“…MSE have advantage to measure the irregularity time series under different time scales 51 . It's designed to reflect the degree of health condition of a biological system by its output physiological signals 53 , which is particularly developed to analyze nonlinear and non-stationary constant signals, and their relationship. It has been hypothesized that as part of complexity physiologic systems, EEG and HRV have similar changing trend on the same time scale 41 .…”

Alzheimer’s Disease and Frontotemporal Dementia: A Robust Classification Method of EEG Signals and a Comparison of Validation Methods

“…In addition, EEG can be used in the differential diagnosis between AD and other diseases leading to dementia, such as vascular brain damage [20,21] and Lewy body dementia [22][23][24][25][26]. Similar methodologies to the one proposed, EEG signal processing and supervised learning classification for recognizing AD patients, have been commonly used in the last decade [27][28][29]. Fiscon et al used Fourier analysis and wavelet transform to extract EEG features which then were classified by a decision trees algorithm [30].…”