Multiway array decomposition analysis of EEGs in Alzheimer's disease

A Comparative Study on Feature Extraction Techniques for the Discrimination of Frontotemporal Dementia and Alzheimer's Disease with Electroencephalography in Resting-State Adults

Early-stage Alzheimer's disease (AD) and frontotemporal dementia (FTD) share similar symptoms, complicating their diagnosis and the development of specific treatment strategies. Our study evaluated multiple feature extraction techniques for identifying AD and FTD biomarkers from electroencephalographic (EEG) signals. We developed an optimised machine learning architecture that integrates sliding windowing, feature extraction, and supervised learning to distinguish between AD and FTD patients, as well as from healthy controls (HCs). Our model, with a 90% overlap for sliding windowing, SVD entropy for feature extraction, and K-Nearest Neighbors (KNN) for supervised learning, achieved a mean F1-score and accuracy of 93% and 91%, 92.5% and 93%, and 91.5% and 91% for discriminating AD and HC, FTD and HC, and AD and FTD, respectively. The feature importance array, an explainable AI feature, highlighted the brain lobes that contributed to identifying and distinguishing AD and FTD biomarkers. This research introduces a novel framework for detecting and discriminating AD and FTD using EEG signals, addressing the need for accurate early-stage diagnostics. Furthermore, a comparative evaluation of sliding windowing, multiple feature extraction, and machine learning methods on AD/FTD detection and discrimination is documented.

A tensor decomposition scheme for EEG-based diagnosis of mild cognitive impairment

Mild Cognitive Impairment (MCI) is the primary stage of acute Alzheimer's disease, and early detection is crucial for the person and those around him. It is difficult to recognize since this mild stage does not have clear clinical signs, and its symptoms are between normal aging and severe dementia. Here, we propose a tensor decomposition-based scheme for automatically diagnosing MCI using Electroencephalogram (EEG) signals. A new projection is proposed, which preserves the spatial information of the electrodes to construct a data tensor. Then, using parallel factor analysis (PARAFAC) tensor decomposition, the features are extracted, and a support vector machine (SVM) is used to discriminate MCI from normal subjects. The proposed scheme was tested on two different datasets. The results showed that the tensor-based method outperformed conventional methods in diagnosing MCI with an average classification accuracy of 93.96% and 78.65% for the first and second datasets, respectively. Therefore, it seems that maintaining the spatial topology of the signals plays a vital role in the processing of EEG signals.

EEG multi-domain feature transfer based on sparse regularized Tucker decomposition

Tensor analysis of electroencephalogram (EEG) can extract the activity information and the potential interaction between different brain regions. However, EEG data varies between subjects, and the existing tensor decomposition algorithms cannot guarantee that the features across subjects are distributed in the same domain, which leads to the non-objectivity of the classification result and analysis, In addition, traditional Tucker decomposition is prone to the explosion of feature dimensions. To solve these problems, combined with the idea of feature transfer, a novel EEG tensor transfer algorithm, Tensor Subspace Learning based on Sparse Regularized Tucker Decomposition (TSL-SRT), is proposed in this paper. In TSL-SRT, new EEG samples are considered as the target domain and original samples as the source domain. The target features can be obtained by projecting the target tensor to the source feature space to ensure that all features are in the same domain. Furthermore, to solve the problem of dimension explosion caused by TSL-SRT, a redundant EEG features screening algorithm is adopted to eliminate the redundant features, and achieves 77.8%, 73.2% and 75.3% accuracy on three BCI datasets. By visualizing the spatial basic matrix of the feature space, it can be seen that TSL-SRT is effective in extracting the features of active brain regions in the BCI task and it can extract the multi-domain features of different subjects in the same domain simultaneously, which provides a new method for the tensor analysis of EEG.

Exploring the feasibility of tensor decomposition for analysis of fNIRS signals: a comparative study with grand averaging method

The analysis of functional near-infrared spectroscopy (fNIRS) signals has not kept pace with the increased use of fNIRS in the behavioral and brain sciences. The popular grand averaging method collapses the oxygenated hemoglobin data within a predefined time of interest window and across multiple channels within a region of interest, potentially leading to a loss of important temporal and spatial information. On the other hand, the tensor decomposition method can reveal patterns in the data without making prior assumptions of the hemodynamic response and without losing temporal and spatial information. The aim of the current study was to examine whether the tensor decomposition method could identify significant effects and novel patterns compared to the commonly used grand averaging method for fNIRS signal analysis. We used two infant fNIRS datasets and applied tensor decomposition (i.e., canonical polyadic and Tucker decompositions) to analyze the significant differences in the hemodynamic response patterns across conditions. The codes are publicly available on GitHub. Bayesian analyses were performed to understand interaction effects. The results from the tensor decomposition method replicated the findings from the grand averaging method and uncovered additional patterns not detected by the grand averaging method. Our findings demonstrate that tensor decomposition is a feasible alternative method for analyzing fNIRS signals, offering a more comprehensive understanding of the data and its underlying patterns.

ETucker: a constrained tensor decomposition for single trial ERP extraction

Objective.In this paper, we propose a new tensor decomposition to extract event-related potentials (ERP) by adding a physiologically meaningful constraint to the Tucker decomposition.Approach.We analyze the performance of the proposed model and compare it with Tucker decomposition by synthesizing a dataset. The simulated dataset is generated using a 12th-order autoregressive model in combination with independent component analysis (ICA) on real no-task electroencephalogram (EEG) recordings. The dataset is manipulated to contain the P300 ERP component and to cover different SNR conditions, ranging from 0 to -30 dB, to simulate the presence of the P300 component in extremely noisy recordings. Furthermore, in order to assess the practicality of the proposed methodology in real-world scenarios, we utilized the brain-computer interface (BCI) competition III-dataset II.Main results.Our primary results demonstrate the superior performance of our approach compared to conventional methods commonly employed for single-trial estimation. Additionally, our method outperformed both Tucker decomposition and non-negative Tucker decomposition in the synthesized dataset. Furthermore, the results obtained from real-world data exhibited meaningful performance and provided insightful interpretations for the extracted P300 component.Significance.The findings suggest that the proposed decomposition is eminently capable of extracting the target P300 component's waveform, including latency and amplitude as well as its spatial location, using single-trial EEG recordings.

Injectable Conductive Hydrogels with Tunable Degradability as Novel Implantable Bioelectrodes

Bioelectrodes have been developed to efficiently mediate electrical signals of biological systems as stimulators and recording devices. Recently, conductive hydrogels have garnered great attention as emerging materials for bioelectrode applications because they can permit intimate/conformal contact with living tissues and tissue-like softness. However, administration and control over the in vivo lifetime of bioelectrodes remain challenges. Here, injectable conductive hydrogels (ICHs) with tunable degradability as implantable bioelectrodes are developed. ICHs were constructed via thiol-ene reactions using poly(ethylene glycol)-tetrathiol and thiol-functionalized reduced graphene oxide with either hydrolyzable poly(ethylene glycol)-diacrylate or stable poly(ethylene glycol)-dimaleimide, the resultant hydrogels of which are degradable and nondegradable, respectively. The ICH electrodes had conductivities of 21-22 mS cm^-1 and Young's moduli of 15-17 kPa, and showed excellent cell and tissue compatibility. The hydrolyzable conductive hydrogels disappeared 3 days after in vivo administration, while the stable conductive hydrogels maintained their shapes for up to 7 days. Our proof-of-concept studies reveal that electromyography signals with significantly improved sensitivity from rats could be obtained from the injected ICH electrodes compared to skin electrodes and injected nonconductive hydrogel electrodes. The ICHs, offering convenience in use, controllable degradation and excellent signal transmission, will have great potential to develop various bioelectronics devices.

Tensor decomposition of TMS-induced EEG oscillations reveals data-driven profiles of antiepileptic drug effects

Transcranial magnetic stimulation combined with electroencephalography is a powerful tool to probe human cortical excitability. The EEG response to TMS stimulation is altered by drugs active in the brain, with characteristic “fingerprints” obtained for drugs of known mechanisms of action. However, the extraction of specific features related to drug effects is not always straightforward as the complex TMS-EEG induced response profile is multi-dimensional. Analytical approaches can rely on a-priori assumptions within each dimension or on the implementation of cluster-based permutations which do not require preselection of specific limits but may be problematic when several experimental conditions are tested. We here propose an alternative data-driven approach based on PARAFAC tensor decomposition, which provides a parsimonious description of the main profiles underlying the multidimensional data. We validated reliability of PARAFAC on TMS-induced oscillations before extracting the features of two common anti-epileptic drugs (levetiracetam and lamotrigine) in an integrated manner. PARAFAC revealed an effect of both drugs, significantly suppressing oscillations in the alpha range in the occipital region. Further, this effect was stronger under the intake of levetiracetam. This study demonstrates, for the first time, that PARAFAC can easily disentangle the effects of subject, drug condition, frequency, time and space in TMS-induced oscillations.

Systematic Review on Resting-State EEG for Alzheimer's Disease Diagnosis and Progression Assessment

Alzheimer's disease (AD) is a neurodegenerative disorder that accounts for nearly 70% of the more than 46 million dementia cases estimated worldwide. Although there is no cure for AD, early diagnosis and an accurate characterization of the disease progression can improve the quality of life of AD patients and their caregivers. Currently, AD diagnosis is carried out using standardized mental status examinations, which are commonly assisted by expensive neuroimaging scans and invasive laboratory tests, thus rendering the diagnosis time consuming and costly. Notwithstanding, over the last decade, electroencephalography (EEG) has emerged as a noninvasive alternative technique for the study of AD, competing with more expensive neuroimaging tools, such as MRI and PET. This paper reports on the results of a systematic review on the utilization of resting-state EEG signals for AD diagnosis and progression assessment. Recent journal articles obtained from four major bibliographic databases were analyzed. A total of 112 journal articles published from January 2010 to February 2018 were meticulously reviewed, and relevant aspects of these papers were compared across articles to provide a general overview of the research on this noninvasive AD diagnosis technique. Finally, recommendations for future studies with resting-state EEG were presented to improve and facilitate the knowledge transfer among research groups.

Spatial-temporal-spectral EEG patterns of BOLD functional network connectivity dynamics

OBJECTIVE

Growing interest in the examination of large-scale brain network functional connectivity dynamics is accompanied by an effort to find the electrophysiological correlates. The commonly used constraints applied to spatial and spectral domains during electroencephalogram (EEG) data analysis may leave part of the neural activity unrecognized. We propose an approach that blindly reveals multimodal EEG spectral patterns that are related to the dynamics of the BOLD functional network connectivity.

APPROACH

The blind decomposition of EEG spectrogram by parallel factor analysis has been shown to be a useful technique for uncovering patterns of neural activity. The simultaneously acquired BOLD fMRI data were decomposed by independent component analysis. Dynamic functional connectivity was computed on the component's time series using a sliding window correlation, and between-network connectivity states were then defined based on the values of the correlation coefficients. ANOVA tests were performed to assess the relationships between the dynamics of between-network connectivity states and the fluctuations of EEG spectral patterns.

MAIN RESULTS

We found three patterns related to the dynamics of between-network connectivity states. The first pattern has dominant peaks in the alpha, beta, and gamma bands and is related to the dynamics between the auditory, sensorimotor, and attentional networks. The second pattern, with dominant peaks in the theta and low alpha bands, is related to the visual and default mode network. The third pattern, also with peaks in the theta and low alpha bands, is related to the auditory and frontal network.

SIGNIFICANCE

Our previous findings revealed a relationship between EEG spectral pattern fluctuations and the hemodynamics of large-scale brain networks. In this study, we suggest that the relationship also exists at the level of functional connectivity dynamics among large-scale brain networks when no standard spatial and spectral constraints are applied on the EEG data.

Multiway Array Decomposition of EEG Spectrum: Implications of Its Stability for the Exploration of Large-Scale Brain Networks

Multiway array decomposition methods have been shown to be promising statistical tools for identifying neural activity in the EEG spectrum. They blindly decompose the EEG spectrum into spatial-temporal-spectral patterns by taking into account inherent relationships among signals acquired at different frequencies and sensors. Our study evaluates the stability of spatial-temporal-spectral patterns derived by one particular method, parallel factor analysis (PARAFAC). We focused on patterns' stability over time and in population and divided the complete data set containing data from 50 healthy subjects into several subsets. Our results suggest that the patterns are highly stable in time, as well as among different subgroups of subjects. Further, we show with simultaneously acquired fMRI data that power fluctuations of some patterns have stable correspondence to hemodynamic fluctuations in large-scale brain networks. We did not find such correspondence for power fluctuations in standard frequency bands, the common way of dealing with EEG data. Altogether, our results suggest that PARAFAC is a suitable method for research in the field of large-scale brain networks and their manifestation in EEG signal.

Using Matrix and Tensor Factorizations for the Single-Trial Analysis of Population Spike Trains

Advances in neuronal recording techniques are leading to ever larger numbers of simultaneously monitored neurons. This poses the important analytical challenge of how to capture compactly all sensory information that neural population codes carry in their spatial dimension (differences in stimulus tuning across neurons at different locations), in their temporal dimension (temporal neural response variations), or in their combination (temporally coordinated neural population firing). Here we investigate the utility of tensor factorizations of population spike trains along space and time. These factorizations decompose a dataset of single-trial population spike trains into spatial firing patterns (combinations of neurons firing together), temporal firing patterns (temporal activation of these groups of neurons) and trial-dependent activation coefficients (strength of recruitment of such neural patterns on each trial). We validated various factorization methods on simulated data and on populations of ganglion cells simultaneously recorded in the salamander retina. We found that single-trial tensor space-by-time decompositions provided low-dimensional data-robust representations of spike trains that capture efficiently both their spatial and temporal information about sensory stimuli. Tensor decompositions with orthogonality constraints were the most efficient in extracting sensory information, whereas non-negative tensor decompositions worked well even on non-independent and overlapping spike patterns, and retrieved informative firing patterns expressed by the same population in response to novel stimuli. Our method showed that populations of retinal ganglion cells carried information in their spike timing on the ten-milliseconds-scale about spatial details of natural images. This information could not be recovered from the spike counts of these cells. First-spike latencies carried the majority of information provided by the whole spike train about fine-scale image features, and supplied almost as much information about coarse natural image features as firing rates. Together, these results highlight the importance of spike timing, and particularly of first-spike latencies, in retinal coding.

Hidden discriminative features extraction for supervised high-order time series modeling

In this paper, an orthogonal Tucker-decomposition-based extraction of high-order discriminative subspaces from a tensor-based time series data structure is presented, named as Tensor Discriminative Feature Extraction (TDFE). TDFE relies on the employment of category information for the maximization of the between-class scatter and the minimization of the within-class scatter to extract optimal hidden discriminative feature subspaces that are simultaneously spanned by every modality for supervised tensor modeling. In this context, the proposed tensor-decomposition method provides the following benefits: i) reduces dimensionality while robustly mining the underlying discriminative features, ii) results in effective interpretable features that lead to an improved classification and visualization, and iii) reduces the processing time during the training stage and the filtering of the projection by solving the generalized eigenvalue issue at each alternation step. Two real third-order tensor-structures of time series datasets (an epilepsy electroencephalogram (EEG) that is modeled as channel×frequency bin×time frame and a microarray data that is modeled as gene×sample×time) were used for the evaluation of the TDFE. The experiment results corroborate the advantages of the proposed method with averages of 98.26% and 89.63% for the classification accuracies of the epilepsy dataset and the microarray dataset, respectively. These performance averages represent an improvement on those of the matrix-based algorithms and recent tensor-based, discriminant-decomposition approaches; this is especially the case considering the small number of samples that are used in practice.

Space-by-time decomposition for single-trial decoding of M/EEG activity

We develop a novel methodology for the single-trial analysis of multichannel time-varying neuroimaging signals. We introduce the space-by-time M/EEG decomposition, based on Non-negative Matrix Factorization (NMF), which describes single-trial M/EEG signals using a set of non-negative spatial and temporal components that are linearly combined with signed scalar activation coefficients. We illustrate the effectiveness of the proposed approach on an EEG dataset recorded during the performance of a visual categorization task. Our method extracts three temporal and two spatial functional components achieving a compact yet full representation of the underlying structure, which validates and summarizes succinctly results from previous studies. Furthermore, we introduce a decoding analysis that allows determining the distinct functional role of each component and relating them to experimental conditions and task parameters. In particular, we demonstrate that the presented stimulus and the task difficulty of each trial can be reliably decoded using specific combinations of components from the identified space-by-time representation. When comparing with a sliding-window linear discriminant algorithm, we show that our approach yields more robust decoding performance across participants. Overall, our findings suggest that the proposed space-by-time decomposition is a meaningful low-dimensional representation that carries the relevant information of single-trial M/EEG signals.

•

Space-by-time decomposition for multichannel time-varying signal analysis.

•

Extraction of spatial and temporal components of single-trial M/EEG activity.

•

Full and succinct characterization of EEG data during a visual categorization task.

•

Single-trial decoding based on task-relevant features.

•

Robust and consistent decoding results across participants.

Epoch-based Entropy for Early Screening of Alzheimer’s Disease

In this paper, we introduce a novel entropy measure, termed epoch-based entropy. This measure quantifies disorder of EEG signals both at the time level and spatial level, using local density estimation by a Hidden Markov Model on inter-channel stationary epochs. The investigation is led on a multi-centric EEG database recorded from patients at an early stage of Alzheimer’s disease (AD) and age-matched healthy subjects. We investigate the classification performances of this method, its robustness to noise, and its sensitivity to sampling frequency and to variations of hyperparameters. The measure is compared to two alternative complexity measures, Shannon’s entropy and correlation dimension. The classification accuracies for the discrimination of AD patients from healthy subjects were estimated using a linear classifier designed on a development dataset, and subsequently tested on an independent test set. Epoch-based entropy reached a classification accuracy of 83% on the test dataset (specificity = 83.3%, sensitivity = 82.3%), outperforming the two other complexity measures. Furthermore, it was shown to be more stable to hyperparameter variations, and less sensitive to noise and sampling frequency disturbances than the other two complexity measures.

A novel biomarker of amnestic MCI based on dynamic cross-frequency coupling patterns during cognitive brain responses

The detection of mild cognitive impairment (MCI), the transitional stage between normal cognitive changes of aging and the cognitive decline caused by AD, is of paramount clinical importance, since MCI patients are at increased risk of progressing into AD. Electroencephalographic (EEG) alterations in the spectral content of brainwaves and connectivity at resting state have been associated with early-stage AD. Recently, cognitive event-related potentials (ERPs) have entered into the picture as an easy to perform screening test. Motivated by the recent findings about the role of cross-frequency coupling (CFC) in cognition, we introduce a relevant methodological approach for detecting MCI based on cognitive responses from a standard auditory oddball paradigm. By using the single trial signals recorded at Pz sensor and comparing the responses to target and non-target stimuli, we first demonstrate that increased CFC is associated with the cognitive task. Then, considering the dynamic character of CFC, we identify instances during which the coupling between particular pairs of brainwave frequencies carries sufficient information for discriminating between normal subjects and patients with MCI. In this way, we form a multiparametric signature of impaired cognition. The new composite biomarker was tested using data from a cohort that consists of 25 amnestic MCI patients and 15 age-matched controls. Standard machine-learning algorithms were employed so as to implement the binary classification task. Based on leave-one-out cross-validation, the measured classification rate was found reaching very high levels (95%). Our approach compares favorably with the traditional alternative of using the morphology of averaged ERP response to make the diagnosis and the usage of features from spectro-temporal analysis of single-trial responses. This further indicates that task-related CFC measurements can provide invaluable analytics in AD diagnosis and prognosis.

A Novel Application of Multiscale Entropy in Electroencephalography to Predict the Efficacy of Acetylcholinesterase Inhibitor in Alzheimer's Disease

Alzheimer's disease (AD) is the most common form of dementia. According to one hypothesis, AD is caused by the reduced synthesis of the neurotransmitter acetylcholine. Therefore, acetylcholinesterase (AChE) inhibitors are considered to be an effective therapy. For clinicians, however, AChE inhibitors are not a predictable treatment for individual patients. We aimed to disclose the difference by biosignal processing. In this study, we used multiscale entropy (MSE) analysis, which can disclose the embedded information in different time scales, in electroencephalography (EEG), in an attempt to predict the efficacy of AChE inhibitors. Seventeen newly diagnosed AD patients were enrolled, with an initial minimental state examination (MMSE) score of 18.8 ± 4.5. After 12 months of AChE inhibitor therapy, 7 patients were responsive and 10 patients were nonresponsive. The major difference between these two groups is Slope 2 (MSE6 to 20). The area below the receiver operating characteristic (ROC) curve of Slope 2 is 0.871 (95% CI = 0.69–1). The sensitivity is 85.7% and the specificity is 60%, whereas the cut-off value of Slope 2 is −0.024. Therefore, MSE analysis of EEG signals, especially Slope 2, provides a potential tool for predicting the efficacy of AChE inhibitors prior to therapy.

Multiscale entropy analysis of resting-state magnetoencephalogram with tensor factorisations in Alzheimer's disease

Tensor factorisations have proven useful to model amplitude and spectral information of brain recordings. Here, we assess the usefulness of tensor factorisations in the multiway analysis of other brain signal features in the context of complexity measures recently proposed to inspect multiscale dynamics. We consider the "refined composite multiscale entropy" (rcMSE), which computes entropy "profiles" showing levels of physiological complexity over temporal scales for individual signals. We compute the rcMSE of resting-state magnetoencephalogram (MEG) recordings from 36 patients with Alzheimer's disease and 26 control subjects. Instead of traditional simple visual examinations, we organise the entropy profiles as a three-way tensor to inspect relationships across temporal and spatial scales and subjects with multiway data analysis techniques based on PARAFAC and PARAFAC2 factorisations. A PARAFAC2 model with two factors was appropriate to account for the interactions in the entropy tensor between temporal scales and MEG channels for all subjects. Moreover, the PARAFAC2 factors had information related to the subjects' diagnosis, achieving a cross-validated area under the ROC curve of 0.77. This confirms the suitability of tensor factorisations to represent electrophysiological brain data efficiently despite the unsupervised nature of these techniques. This article is part of a Special Issue entitled 'Neural data analysis'.

A novel method of early diagnosis of Alzheimer's disease based on EEG signals

Studies have reported that electroencephalogram signals in Alzheimer's disease patients usually have less synchronization than those of healthy subjects. Changes in electroencephalogram signals start at early stage but, clinically, these changes are not easily detected. To detect this perturbation, three neural synchrony measurement techniques: phase synchrony, magnitude squared coherence, and cross correlation are applied to three different databases of mild Alzheimer's disease patients and healthy subjects. We have compared the right and left temporal lobes of the brain with the rest of the brain areas (frontal, central, and occipital) as temporal regions are relatively the first ones to be affected by Alzheimer's disease. Moreover, electroencephalogram signals are further classified into five different frequency bands (delta, theta, alpha beta, and gamma) because each frequency band has its own physiological significance in terms of signal evaluation. A new approach using principal component analysis before applying neural synchrony measurement techniques has been presented and compared with Average technique. The simulation results indicated that applying principal component analysis before synchrony measurement techniques shows significantly better results as compared to the lateral one. At the end, all the aforementioned techniques are assessed by a statistical test (Mann-Whitney U test) to compare the results.

BENEFITS OF MULTI-DOMAIN FEATURE OF MISMATCH NEGATIVITY EXTRACTED BY NON-NEGATIVE TENSOR FACTORIZATION FROM EEG COLLECTED BY LOW-DENSITY ARRAY

Through exploiting temporal, spectral, time-frequency representations, and spatial properties of mismatch negativity (MMN) simultaneously, this study extracts a multi-domain feature of MMN mainly using non-negative tensor factorization. In our experiment, the peak amplitude of MMN between children with reading disability and children with attention deficit was not significantly different, whereas the new feature of MMN significantly discriminated the two groups of children. This is because the feature was derived from multi-domain information with significant reduction of the heterogeneous effect of datasets.

Empirical mode decomposition based detrended sample entropy in electroencephalography for Alzheimer's disease

Quantitative electroencephalographs (qEEG) provide a potential method to objectively quantify the cortical activations in Alzheimer's disease (AD), but they are too insensitive to probe the alteration of EEG in the early AD. The sample entropy (SaEn) attempts to quantify the complex information embedded in EEG non-linearly, which fits in that EEG originates from non-linear interactions. However, a technical issue which has been ignored by most researchers is that the signal should be stationary. In order to resolve the non-stationarity of SaEn in EEG to improve the sensitivity, an empirical mode decomposition (EMD) was applied for detrending in this study. Twenty-seven AD patients (9M/18F; mean age 74.0±1.5 years) were included. Their initial Minimal Mental Status Examination was 19.3±0.7. They received the first resting-awake 30-mine EEG before the therapy. Five of them received a follow-up examination within 6 months after the therapy. The 30-s EEG data without artifacts were selected and analyzed with a new proposed method, "EMD-based detrended-SaEn" to attenuate the influence of intrinsic non-stationarity. The correlation factors in 27 AD patients showed a moderate correlation (0.361-0.523, p<0.05) between MMSE and EMD-based detrended SaEn in Fp1, Fp2, F4 and T3. There was a high correlation (Correlation coefficient=0.975, p<0.05) between the changes of MMSE and the changes of EMD-based detrended-SaEn in F7 in 5 follow-up patients. The dynamic complexity of EEG fluctuations is degraded by pathological degeneration, and EMD-based detrended SaEn provides an objective, non-invasive and non-expensive tool for evaluating and following AD patients.