Imaging of Nonlinear and Dynamic Functional Brain Connectivity Based on EEG Recordings With the Application on the Diagnosis of Alzheimer's Disease

Functional and effective EEG connectivity patterns in Alzheimer's disease and mild cognitive impairment: a systematic review

Background

Alzheimer's disease (AD) might be best conceptualized as a disconnection syndrome, such that symptoms may be largely attributable to disrupted communication between brain regions, rather than to deterioration within discrete systems. EEG is uniquely capable of directly and non-invasively measuring neural activity with precise temporal resolution; connectivity quantifies the relationships between such signals in different brain regions. EEG research on connectivity in AD and mild cognitive impairment (MCI), often considered a prodromal phase of AD, has produced mixed results and has yet to be synthesized for comprehensive review. Thus, we performed a systematic review of EEG connectivity in MCI and AD participants compared with cognitively healthy older adult controls.

Methods

We searched PsycINFO, PubMed, and Web of Science for peer-reviewed studies in English on EEG, connectivity, and MCI/AD relative to controls. Of 1,344 initial matches, 124 articles were ultimately included in the systematic review.

Results

The included studies primarily analyzed coherence, phase-locked, and graph theory metrics. The influence of factors such as demographics, design, and approach was integrated and discussed. An overarching pattern emerged of lower connectivity in both MCI and AD compared to healthy controls, which was most prominent in the alpha band, and most consistent in AD. In the minority of studies reporting greater connectivity, theta band was most commonly implicated in both AD and MCI, followed by alpha. The overall prevalence of alpha effects may indicate its potential to provide insight into nuanced changes associated with AD-related networks, with the caveat that most studies were during the resting state where alpha is the dominant frequency. When greater connectivity was reported in MCI, it was primarily during task engagement, suggesting compensatory resources may be employed. In AD, greater connectivity was most common during rest, suggesting compensatory resources during task engagement may already be exhausted.

Conclusion

The review highlighted EEG connectivity as a powerful tool to advance understanding of AD-related changes in brain communication. We address the need for including demographic and methodological details, using source space connectivity, and extending this work to cognitively healthy older adults with AD risk toward advancing early AD detection and intervention.

Dementia rhythms: Unveiling the EEG dynamics for MCI detection through spectral and synchrony neuromarkers

BACKGROUND

Neurological disorders arise primarily from the dysfunction of brain cells, leading to various impairments. Electroencephalography (EEG) stands out as the most popular method in the discovery of neuromarkers indicating neurological disorders. The proposed study investigates the effectiveness of spectral and synchrony neuromarkers derived from resting state EEG in the detection of Mild Cognitive Impairment (MCI) with controls.

NEW METHODS

The dataset is composed of 10 MCI and 10 HC groups. Spectral features and synchrony measures are utilized to detect slowing patterns in MCI. Efficient neuro-markers are classified by 25 classification algorithm. Independent samples t-test and Pearson's Correlation Coefficients are applied to reveal group differences for spectral markers, and repeated measures ANOVA is tested for wPLI-based markers.

RESULTS

Lower peak amplitudes are prominent in MCI participants for high frequencies indicating slower physiological behavior of the demented EEG. The MCI and HC groups are correctly classified with 95 % acc. using peak amplitudes of beta band with LGBM classifier. Higher wPLI values are calculated for HC participants in high frequencies. The alpha wPLI values achieve a classification accuracy of 99 % using the LGBM algorithm for MCI detection.

COMPARISON WITH EXISTING METHODS

The neuro-markers including peak amplitudes, frequencies, and wPLIs with advanced machine learning techniques showcases the innovative nature of this research.

CONCLUSION

The findings suggest that peak amplitudes and wPLI in high frequency bands derived from resting state EEG are effective neuromarkers for detection of MCI. Spectral and synchrony neuro-markers hold great promise for accurate MCI detection.

HiRENet: Novel convolutional neural network architecture using Hilbert-transformed and raw electroencephalogram (EEG) for subject-independent emotion classification

BACKGROUND AND OBJECTIVES

Convolutional neural networks (CNNs) are the most widely used deep-learning framework for decoding electroencephalograms (EEGs) due to their exceptional ability to extract hierarchical features from high-dimensional EEG data. Traditionally, CNNs have primarily utilized multi-channel raw EEG data as the input tensor; however, the performance of CNN-based EEG decoding may be enhanced by incorporating phase information alongside amplitude information.

METHODS

This study introduces a novel CNN architecture called the Hilbert-transformed (HT) and raw EEG network (HiRENet), which incorporates both raw and HT EEG as inputs. This concurrent use of HT and raw EEG aims to integrate phase information with existing amplitude information, potentially offering a more comprehensive reflection of functional connectivity across various brain regions. The HiRENet model was developed using two CNN frameworks: ShallowFBCSPNet and a CNN with a residual block (ResCNN). The performance of the HiRENet model was assessed using a lab-made EEG database to classify human emotions, comparing three input modalities: raw EEG, HT EEG, and a combination of both signals. Additionally, the computational complexity was evaluated to validate the computational efficiency of the ResCNN design.

RESULTS

The HiRENet model based on ResCNN achieved the highest classification accuracy, with 86.03% for valence and 84.01% for arousal classifications, surpassing traditional CNN methodologies. Considering computational efficiency, ResCNN demonstrated superiority over ShallowFBCSPNet in terms of speed and inference time, despite having a higher parameter count.

CONCLUSION

Our experimental results showed that the proposed HiRENet can be potentially used as a new option to improve the overall performance for deep learning-based EEG decoding problems.

A comprehensive survey of complex brain network representation

Recent years have shown great merits in utilizing neuroimaging data to understand brain structural and functional changes, as well as its relationship to different neurodegenerative diseases and other clinical phenotypes. Brain networks, derived from different neuroimaging modalities, have attracted increasing attention due to their potential to gain system-level insights to characterize brain dynamics and abnormalities in neurological conditions. Traditional methods aim to pre-define multiple topological features of brain networks and relate these features to different clinical measures or demographical variables. With the enormous successes in deep learning techniques, graph learning methods have played significant roles in brain network analysis. In this survey, we first provide a brief overview of neuroimaging-derived brain networks. Then, we focus on presenting a comprehensive overview of both traditional methods and state-of-the-art deep-learning methods for brain network mining. Major models, and objectives of these methods are reviewed within this paper. Finally, we discuss several promising research directions in this field.

Fourier Synchrosqueezing Transform-ICA-EMD Framework Based EOG-Biometric Sustainable and Continuous Authentication via Voluntary Eye Blinking Activities

In recent years, limited works on EOG (electrooculography)-based biometric authentication systems have been carried out with eye movements or eye blinking activities in the current literature. EOGs have permanent and unique traits that can separate one individual from another. In this work, we have investigated FSST (Fourier Synchrosqueezing Transform)-ICA (Independent Component Analysis)-EMD (Empirical Mode Decomposition) robust framework-based EOG-biometric authentication (one-versus-others verification) performances using ensembled RNN (Recurrent Neural Network) deep models voluntary eye blinkings movements. FSST is implemented to provide accurate and dense temporal-spatial properties of EOGs on the state-of-the-art time-frequency matrix. ICA is a powerful statistical tool to decompose multiple recording electrodes. Finally, EMD is deployed to isolate EOG signals from the EEGs collected from the scalp. As our best knowledge, this is the first research attempt to explore the success of the FSST-ICA-EMD framework on EOG-biometric authentication generated via voluntary eye blinking activities in the limited EOG-related biometric literature. According to the promising results, improved and high recognition accuracies (ACC/Accuracy: ≥99.99% and AUC/Area under the Curve: 0.99) have been achieved in addition to the high TAR (true acceptance rate) scores (≥98%) and low FAR (false acceptance rate) scores (≤3.33%) in seven individuals. On the other hand, authentication and monitoring for online users/students are becoming essential and important tasks due to the increase of the digital world (e-learning, e-banking, or e-government systems) and the COVID-19 pandemic. Especially in order to ensure reliable access, a highly scalable and affordable approach for authenticating the examinee without cheating or monitoring high-data-size video streaming is required in e-learning platforms and online education strategies. Hence, this work may present an approach that offers a sustainable, continuous, and reliable EOG-biometric authentication of digital applications, including e-learning platforms for users/students.

Brain network analysis for the discrimination of dementia disorders using electrophysiology signals: A systematic review

Background

Dementia-related disorders have been an age-long challenge to the research and healthcare communities as their various forms are expressed with similar clinical symptoms. These disorders are usually irreversible at their late onset, hence their lack of validated and approved cure. Since their prodromal stages usually lurk for a long period of time before the expression of noticeable clinical symptoms, a secondary prevention which has to do with treating the early onsets has been suggested as the possible solution. Connectivity analysis of electrophysiology signals has played significant roles in the diagnosis of various dementia disorders through early onset identification.

Objective

With the various applications of electrophysiology signals, the purpose of this study is to systematically review the step-by-step procedures of connectivity analysis frameworks for dementia disorders. This study aims at identifying the methodological issues involved in such frameworks and also suggests approaches to solve such issues.

Methods

In this study, ProQuest, PubMed, IEEE Xplore, Springer Link, and Science Direct databases are employed for exploring the evolution and advancement of connectivity analysis of electrophysiology signals of dementia-related disorders between January 2016 to December 2022. The quality of assessment of the studied articles was done using Cochrane guidelines for the systematic review of diagnostic test accuracy.

Results

Out of a total of 4,638 articles found to have been published on the review scope between January 2016 to December 2022, a total of 51 peer-review articles were identified to completely satisfy the review criteria. An increasing trend of research in this domain is identified within the considered time frame. The ratio of MEG and EEG utilization found within the reviewed articles is 1:8. Most of the reviewed articles employed graph theory metrics for their analysis with clustering coefficient (CC), global efficiency (GE), and characteristic path length (CPL) appearing more frequently compared to other metrics.

Significance

This study provides general insight into how to employ connectivity measures for the analysis of electrophysiology signals of dementia-related disorders in order to better understand their underlying mechanism and their differential diagnosis.

EEG/fNIRS Based Workload Classification Using Functional Brain Connectivity and Machine Learning

There is high demand for techniques to estimate human mental workload during some activities for productivity enhancement or accident prevention. Most studies focus on a single physiological sensing modality and use univariate methods to analyse multi-channel electroencephalography (EEG) data. This paper proposes a new framework that relies on the features of hybrid EEG-functional near-infrared spectroscopy (EEG-fNIRS), supported by machine-learning features to deal with multi-level mental workload classification. Furthermore, instead of the well-used univariate power spectral density (PSD) for EEG recording, we propose using bivariate functional brain connectivity (FBC) features in the time and frequency domains of three bands: delta (0.5-4 Hz), theta (4-7 Hz) and alpha (8-15 Hz). With the assistance of the fNIRS oxyhemoglobin and deoxyhemoglobin (HbO and HbR) indicators, the FBC technique significantly improved classification performance at a 77% accuracy for 0-back vs. 2-back and 83% for 0-back vs. 3-back using a public dataset. Moreover, topographic and heat-map visualisation indicated that the distinguishing regions for EEG and fNIRS showed a difference among the 0-back, 2-back and 3-back test results. It was determined that the best region to assist the discrimination of the mental workload for EEG and fNIRS is different. Specifically, the posterior area performed the best for the posterior midline occipital (POz) EEG in the alpha band and fNIRS had superiority in the right frontal region (AF8).

Efficient graph convolutional networks for seizure prediction using scalp EEG

Epilepsy is a chronic brain disease that causes persistent and severe damage to the physical and mental health of patients. Daily effective prediction of epileptic seizures is crucial for epilepsy patients especially those with refractory epilepsy. At present, a large number of deep learning algorithms such as Convolutional Neural Networks and Recurrent Neural Networks have been used to predict epileptic seizures and have obtained better performance than traditional machine learning methods. However, these methods usually transform the Electroencephalogram (EEG) signal into a Euclidean grid structure. The conversion suffers from loss of adjacent spatial information, which results in deep learning models requiring more storage and computational consumption in the process of information fusion after information extraction. This study proposes a general Graph Convolutional Networks (GCN) model architecture for predicting seizures to solve the problem of oversized seizure prediction models based on exploring the graph structure of EEG signals. As a graph classification task, the network architecture includes graph convolution layers that extract node features with one-hop neighbors, pooling layers that summarize abstract node features; and fully connected layers that implement classification, resulting in superior prediction performance and smaller network size. The experiment shows that the model has an average sensitivity of 96.51%, an average AUC of 0.92, and a model size of 15.5 k on 18 patients in the CHB-MIT scalp EEG dataset. Compared with traditional deep learning methods, which require a large number of parameters and computational effort and are demanding in terms of storage space and energy consumption, this method is more suitable for implementation on compact, low-power wearable devices as a standard process for building a generic low-consumption graph network model on similar biomedical signals. Furthermore, the edge features of graphs can be used to make a preliminary determination of locations and types of discharge, making it more clinically interpretable.

Multimodal explainable AI predicts upcoming speech behavior in adults who stutter

A key goal of cognitive neuroscience is to better understand how dynamic brain activity relates to behavior. Such dynamics, in terms of spatial and temporal patterns of brain activity, are directly measured with neurophysiological methods such as EEG, but can also be indirectly expressed by the body. Autonomic nervous system activity is the best-known example, but, muscles in the eyes and face can also index brain activity. Mostly parallel lines of artificial intelligence research show that EEG and facial muscles both encode information about emotion, pain, attention, and social interactions, among other topics. In this study, we examined adults who stutter (AWS) to understand the relations between dynamic brain and facial muscle activity and predictions about future behavior (fluent or stuttered speech). AWS can provide insight into brain-behavior dynamics because they naturally fluctuate between episodes of fluent and stuttered speech behavior. We focused on the period when speech preparation occurs, and used EEG and facial muscle activity measured from video to predict whether the upcoming speech would be fluent or stuttered. An explainable self-supervised multimodal architecture learned the temporal dynamics of both EEG and facial muscle movements during speech preparation in AWS, and predicted fluent or stuttered speech at 80.8% accuracy (chance=50%). Specific EEG and facial muscle signals distinguished fluent and stuttered trials, and systematically varied from early to late speech preparation time periods. The self-supervised architecture successfully identified multimodal activity that predicted upcoming behavior on a trial-by-trial basis. This approach could be applied to understanding the neural mechanisms driving variable behavior and symptoms in a wide range of neurological and psychiatric disorders. The combination of direct measures of neural activity and simple video data may be applied to developing technologies that estimate brain state from subtle bodily signals.

Ultra-high-resolution time-frequency analysis of EEG to characterise brain functional connectivity with the application in Alzheimer's disease

OBJECTIVE

This study aims to explore the potential of high-resolution brain functional connectivity based on electroencephalogram (EEG), a non-invasive low-cost technique, to be translated into a long-overdue biomarker and a diagnostic method for Alzheimer's disease (AD).

APPROACH

The paper proposes a novel ultra-high-resolution time-frequency nonlinear cross-spectrum method to construct a promising biomarker of AD pathophysiology. Specifically, using the peak frequency estimated from a Revised Hilbert-Huang Transformation cross-spectrum as a biomarker, the Support Vector Machine classifier is used to distinguish AD from healthy controls (HC).

MAIN RESULTS

With the combinations of the proposed biomarker and machine learning, we achieved a promising accuracy of 89%. The proposed method performs better than the wavelet cross-spectrum and other functional connectivity measures in the temporal or frequency domain, particularly in the Full, Delta and Alpha bands. Besides, a novel visualisation approach developed from topography is introduced to represent the brain functional connectivity, with which the difference between AD and HCs can be clearly displayed. The interconnections between posterior and other brain regions are obviously affected in AD.

SIGNIFICANCE

Those findings imply that the proposed RHHT approach could better track dynamic and nonlinear functional connectivity information, paving the way for the development of a novel diagnostic approach.

Spatial-temporal graph convolutional network for Alzheimer classification based on brain functional connectivity imaging of electroencephalogram

Functional connectivity of the human brain, representing statistical dependence of information flow between cortical regions, significantly contributes to the study of the intrinsic brain network and its functional mechanism. To fully explore its potential in the early diagnosis of Alzheimer's disease (AD) using electroencephalogram (EEG) recordings, this article introduces a novel dynamical spatial-temporal graph convolutional neural network (ST-GCN) for better classification performance. Different from existing studies that are based on either topological brain function characteristics or temporal features of EEG, the proposed ST-GCN considers both the adjacency matrix of functional connectivity from multiple EEG channels and corresponding dynamics of signal EEG channel simultaneously. Different from the traditional graph convolutional neural networks, the proposed ST-GCN makes full use of the constrained spatial topology of functional connectivity and the discriminative dynamic temporal information represented by the 1D convolution. We conducted extensive experiments on the clinical EEG data set of AD patients and Healthy Controls. The results demonstrate that the proposed method achieves better classification performance (92.3%) than the state-of-the-art methods. This approach can not only help diagnose AD but also better understand the effect of normal ageing on brain network characteristics before we can accurately diagnose the condition based on resting-state EEG.

EEG Signal Processing and Supervised Machine Learning to Early Diagnose Alzheimer’s Disease

Electroencephalography (EEG) signal analysis is a fast, inexpensive, and accessible technique to detect the early stages of dementia, such as Mild Cognitive Impairment (MCI) and Alzheimer’s disease (AD). In the last years, EEG signal analysis has become an important topic of research to extract suitable biomarkers to determine the subject’s cognitive impairment. In this work, we propose a novel simple and efficient method able to extract features with a finite response filter (FIR) in the double time domain in order to discriminate among patients affected by AD, MCI, and healthy controls (HC). Notably, we compute the power intensity for each high- and low-frequency band, using their absolute differences to distinguish among the three classes of subjects by means of different supervised machine learning methods. We use EEG recordings from a cohort of 105 subjects (48 AD, 37 MCI, and 20 HC) referred for dementia to the IRCCS Centro Neurolesi “Bonino-Pulejo” of Messina, Italy. The findings show that this method reaches 97%, 95%, and 83% accuracy when considering binary classifications (HC vs. AD, HC vs. MCI, and MCI vs. AD) and an accuracy of 75% when dealing with the three classes (HC vs. AD vs. MCI). These results improve upon those obtained in previous studies and demonstrate the validity of our approach. Finally, the efficiency of the proposed method might allow its future development on embedded devices for low-cost real-time diagnosis.

qEEG Analysis in the Diagnosis of Alzheimer’s Disease: A Comparison of Functional Connectivity and Spectral Analysis

Alzheimer’s disease (AD) is a brain disorder that is mainly characterized by a progressive degeneration of neurons in the brain and decline of cognitive abilities. This study compared an FFT-based spectral analysis against a functional connectivity analysis for the diagnosis of AD. Both quantitative methods were applied on an EEG dataset including 20 diagnosed AD patients and 20 age-matched healthy controls (HC). The obtained results showed an advantage of the functional connectivity analysis when compared to the spectral analysis; while the latter could not find any significant differences between the AD and HC groups, the functional connectivity analysis showed statistically higher synchronization levels in the AD group in the lower frequency bands (delta and theta), suggesting a ‘phase-locked’ state in AD-affected brains. Further comparison of functional connectivity between the homotopic regions confirmed that the traits of AD were localized to the centro-parietal and centro-temporal areas in the theta frequency band (4–8 Hz). This study applies a neural metric for Alzheimer’s detection from a data science perspective rather than from a neuroscience one and shows that the combination of bipolar derivations with phase synchronization yields similar results to comparable studies employing alternative analysis methods.

Exploring dynamic functional connectivity alterations in the preclinical stage of Alzheimer's disease: an exploratory study from SILCODE

INTRODUCTION

Exploring functional connectivity (FC) alterations is important for the understanding of underlying neuronal network alterations in subjective cognitive decline (SCD). The objective of this study was to prove that dynamic FC can better reflect the changes of brain function in individuals with SCD compared to static FC, and further to explore the association between FC alterations and amyloid pathology in the preclinical stage of Alzheimer's disease (AD).

METHODS

101 normal control (NC) subjects, 97 SCDs, and 55 cognitive impairment (CI) subjects constituted the whole-cohort. Of these, 29 NCs and 52 SCDs with amyloid images were selected as the sub-cohort. First, independent components (ICs) were identified by independent component analysis and static and dynamic FC were calculated by pairwise correlation coefficient between ICs. Second, FC alterations were identified through group comparison, and seed-based dynamic FC analysis was done. Analysis of variance (ANOVA) was used to compare the seed-based dynamic FC maps and measure the group or amyloid effects. Finally, correlation analysis was conducted between the altered dynamic FC and amyloid burden.

RESULTS

The results showed that 42 ICs were revealed. Significantly altered dynamic FC included those between the salience/ventral attention network, the default mode network, and the visual network. Specifically, the thalamus/caudate (IC 25) drove the hub role in the group differences. In the seed-based dynamic FC analysis, the dynamic FC between the thalamus/caudate and the middle temporal/frontal gyrus was observed to be higher in the SCD and CI groups. Moreover, a higher dynamic FC between the thalamus/caudate and visual cortex was observed in the amyloid positive group. Finally, the altered dynamic FC was associated with the amyloid global standardized uptake value ratio (SUVr).

CONCLUSION

Our findings suggest SCD-related alterations could be more reflected by dynamic FC than static FC, and the alterations are associated with global SUVr.

Brain functional and effective connectivity based on electroencephalography recordings: A review

Functional connectivity and effective connectivity of the human brain, representing statistical dependence and directed information flow between cortical regions, significantly contribute to the study of the intrinsic brain network and its functional mechanism. Many recent studies on electroencephalography (EEG) have been focusing on modeling and estimating brain connectivity due to increasing evidence that it can help better understand various brain neurological conditions. However, there is a lack of a comprehensive updated review on studies of EEG-based brain connectivity, particularly on visualization options and associated machine learning applications, aiming to translate those techniques into useful clinical tools. This article reviews EEG-based functional and effective connectivity studies undertaken over the last few years, in terms of estimation, visualization, and applications associated with machine learning classifiers. Methods are explored and discussed from various dimensions, such as either linear or nonlinear, parametric or nonparametric, time-based, and frequency-based or time-frequency-based. Then it is followed by a novel review of brain connectivity visualization methods, grouped by Heat Map, data statistics, and Head Map, aiming to explore the variation of connectivity across different brain regions. Finally, the current challenges of related research and a roadmap for future related research are presented.

A Triple-Network Dynamic Connection Study in Alzheimer's Disease

Alzheimer's disease (AD) was associated with abnormal organization and function of large-scale brain networks. We applied group independent component analysis (Group ICA) to construct the triple-network consisting of the saliency network (SN), the central executive network (CEN), and the default mode network (DMN) in 25 AD, 60 mild cognitive impairment (MCI) and 60 cognitively normal (CN) subjects. To explore the dynamic functional network connectivity (dFNC), we investigated dynamic time-varying triple-network interactions in subjects using Group ICA analysis based on k-means clustering (GDA-k-means). The mean of brain state-specific network interaction indices (meanNII) in the three groups (AD, MCI, CN) showed significant differences by ANOVA analysis. To verify the robustness of the findings, a support vector machine (SVM) was taken meanNII, gender and age as features to classify. This method obtained accuracy values of 95, 94, and 77% when classifying AD vs. CN, AD vs. MCI, and MCI vs. CN, respectively. In our work, the findings demonstrated that the dynamic characteristics of functional interactions of the triple-networks contributed to studying the underlying pathophysiology of AD. It provided strong evidence for dysregulation of brain dynamics of AD.

Rapid three-dimensional multiparametric MRI with quantitative transient-state imaging

Novel methods for quantitative, transient-state multiparametric imaging are increasingly being demonstrated for assessment of disease and treatment efficacy. Here, we build on these by assessing the most common Non-Cartesian readout trajectories (2D/3D radials and spirals), demonstrating efficient anti-aliasing with a k-space view-sharing technique, and proposing novel methods for parameter inference with neural networks that incorporate the estimation of proton density. Our results show good agreement with gold standard and phantom references for all readout trajectories at 1.5 T and 3 T. Parameters inferred with the neural network were within 6.58% difference from the parameters inferred with a high-resolution dictionary. Concordance correlation coefficients were above 0.92 and the normalized root mean squared error ranged between 4.2 and 12.7% with respect to gold-standard phantom references for T1 and T2. In vivo acquisitions demonstrate sub-millimetric isotropic resolution in under five minutes with reconstruction and inference times < 7 min. Our 3D quantitative transient-state imaging approach could enable high-resolution multiparametric tissue quantification within clinically acceptable acquisition and reconstruction times.