Brain health in diverse settings: How age, demographics and cognition shape brain function.

Diversity in brain health is influenced by individual differences in demographics and cognition. However, most studies on brain health and diseases have typically controlled for these factors rather than explored their potential to predict brain signals. Here, we assessed the role of individual differences in demographics (age, sex, and education; n = 1298) and cognition (n = 725) as predictors of different metrics usually used in case-control studies. These included power spectrum and aperiodic (1/f slope, knee, offset) metrics, as well as complexity (fractal dimension estimation, permutation entropy, Wiener entropy, spectral structure variability) and connectivity (graph-theoretic mutual information, conditional mutual information, organizational information) from the source space resting-state EEG activity in a diverse sample from the global south and north populations. Brain-phenotype models were computed using EEG metrics reflecting local activity (power spectrum and aperiodic components) and brain dynamics and interactions (complexity and graph-theoretic measures). Electrophysiological brain dynamics were modulated by individual differences despite the varied methods of data acquisition and assessments across multiple centers, indicating that results were unlikely to be accounted for by methodological discrepancies. Variations in brain signals were mainly influenced by age and cognition, while education and sex exhibited less importance. Power spectrum activity and graph-theoretic measures were the most sensitive in capturing individual differences. Older age, poorer cognition, and being male were associated with reduced alpha power, whereas older age and less education were associated with reduced network integration and segregation. Findings suggest that basic individual differences impact core metrics of brain function that are used in standard case-control studies. Considering individual variability and diversity in global settings would contribute to a more tailored understanding of brain function.

Nonictal electroencephalographic measures for the diagnosis of functional seizures.

OBJECTIVE: Functional seizures (FS) look like epileptic seizures but are characterized by a lack of epileptic activity in the brain. Approximately one in five referrals to epilepsy clinics are diagnosed with this condition. FS are diagnosed by recording a seizure using video-electroencephalography (EEG), from which an expert inspects the semiology and the EEG. However, this method can be expensive and inaccessible and can present significant patient burden. No single biomarker has been found to diagnose FS. However, the current limitations in FS diagnosis could be improved with machine learning to classify signal features extracted from EEG, thus providing a potentially very useful aid to clinicians. METHODS: The current study has investigated the use of seizure-free EEG signals with machine learning to identify subjects with FS from those with epilepsy. The dataset included interictal and preictal EEG recordings from 48 subjects with FS (mean age = 34.76 ± 10.55 years, 14 males) and 29 subjects with epilepsy (mean age = 38.95 ± 13.93 years, 18 males) from which various statistical, temporal, and spectral features from the five EEG frequency bands were extracted then analyzed with threshold accuracy, five machine learning classifiers, and two feature importance approaches. RESULTS: The highest classification accuracy reported from thresholding was 60.67%. However, the temporal features were the best performing, with the highest balanced accuracy reported by the machine learning models: 95.71% with all frequency bands combined and a support vector machine classifier. SIGNIFICANCE: Machine learning was much more effective than using individual features and could be a powerful aid in FS diagnosis. Furthermore, combining the frequency bands improved the accuracy of the classifiers in most cases, and the lowest performing EEG bands were consistently delta and gamma.

Daily Pain Severity but Not Vertebral Fractures Is Associated With Lower Physical Activity in Postmenopausal Women With Back Pain.

Back pain lifetime incidence is 60%-70%, while 12%-20% of older women have vertebral fractures (VFs), often with back pain. We aimed to provide objective evidence, currently lacking, regarding whether back pain and VFs affect physical activity (PA). We recruited 69 women with recent back pain (age 74.5 ± 5.4 years). Low- (0.5 < g < 1.0), medium- (1.0 ≤ g < 1.5), and high-impact (g ≥ 1.5) PA and walking time were measured (100 Hz for 7 days, hip-worn accelerometer). Linear mixed-effects models assessed associations between self-reported pain and PA, and group differences (VFs from spine radiographs/no-VF) in PA. Higher daily pain was associated with reduced low (ß = -0.12, 95% confidence interval, [-0.22, -0.03], p = .013) and medium-impact PA (ß = -0.11, 95% confidence interval, [-0.21, -0.01], p = .041), but not high-impact PA or walking time (p > .11). VFs were not associated with PA (all p > .2). Higher daily pain levels but not VFs were associated with reduced low- and medium-impact PA, which could increase sarcopenia and falls risk in older women with back pain.

Machine Learning Models for Weight-Bearing Activity Type Recognition Based on Accelerometry in Postmenopausal Women.

Hip-worn triaxial accelerometers are widely used to assess physical activity in terms of energy expenditure. Methods for classification in terms of different types of activity of relevance to the skeleton in populations at risk of osteoporosis are not currently available. This publication aims to assess the accuracy of four machine learning models on binary (standing and walking) and tertiary (standing, walking, and jogging) classification tasks in postmenopausal women. Eighty women performed a shuttle test on an indoor track, of which thirty performed the same test on an indoor treadmill. The raw accelerometer data were pre-processed, converted into eighteen different features and then combined into nine unique feature sets. The four machine learning models were evaluated using three different validation methods. Using the leave-one-out validation method, the highest average accuracy for the binary classification model, 99.61%, was produced by a k-NN Manhattan classifier using a basic statistical feature set. For the tertiary classification model, the highest average accuracy, 94.04%, was produced by a k-NN Manhattan classifier using a feature set that included all 18 features. The methods and classifiers within this study can be applied to accelerometer data to more accurately characterize weight-bearing activity which are important to skeletal health.

Entropy Measures of Electroencephalograms towards the Diagnosis of Psychogenic Non-Epileptic Seizures.

Psychogenic non-epileptic seizures (PNES) may resemble epileptic seizures but are not caused by epileptic activity. However, the analysis of electroencephalogram (EEG) signals with entropy algorithms could help identify patterns that differentiate PNES and epilepsy. Furthermore, the use of machine learning could reduce the current diagnosis costs by automating classification. The current study extracted the approximate sample, spectral, singular value decomposition, and Renyi entropies from interictal EEGs and electrocardiograms (ECG)s of 48 PNES and 29 epilepsy subjects in the broad, delta, theta, alpha, beta, and gamma frequency bands. Each feature-band pair was classified by a support vector machine (SVM), k-nearest neighbour (kNN), random forest (RF), and gradient boosting machine (GBM). In most cases, the broad band returned higher accuracy, gamma returned the lowest, and combining the six bands together improved classifier performance. The Renyi entropy was the best feature and returned high accuracy in every band. The highest balanced accuracy, 95.03%, was obtained by the kNN with Renyi entropy and combining all bands except broad. This analysis showed that entropy measures can differentiate between interictal PNES and epilepsy with high accuracy, and improved performances indicate that combining bands is an effective improvement for diagnosing PNES from EEGs and ECGs.

Sex Differences in the Complexity of Healthy Older Adults' Magnetoencephalograms.

The analysis of resting-state brain activity recording in magnetoencephalograms (MEGs) with new algorithms of symbolic dynamics analysis could help obtain a deeper insight into the functioning of the brain and identify potential differences between males and females. Permutation Lempel-Ziv complexity (PLZC), a recently introduced non-linear signal processing algorithm based on symbolic dynamics, was used to evaluate the complexity of MEG signals in source space. PLZC was estimated in a broad band of frequencies (2-45 Hz), as well as in narrow bands (i.e., theta (4-8 Hz), alpha (8-12 Hz), low beta (12-20 Hz), high beta (20-30 Hz), and gamma (30-45 Hz)) in a sample of 98 healthy elderly subjects (49 males, 49 female) aged 65-80 (average age of 72.71 ± 4.22 for males and 72.67 ± 4.21 for females). PLZC was significantly higher for females than males in the high beta band at posterior brain regions including the precuneus, and the parietal and occipital cortices. Further statistical analyses showed that higher complexity values over highly overlapping regions than the ones mentioned above were associated with larger hippocampal volumes only in females. These results suggest that sex differences in healthy aging can be identified from the analysis of magnetoencephalograms with novel signal processing methods.

Fuzzy Entropy Analysis of the Electroencephalogram in Patients with Alzheimer's Disease: Is the Method Superior to Sample Entropy?

Alzheimer's disease (AD) is the most prevalent form of dementia in the world, which is characterised by the loss of neurones and the build-up of plaques in the brain, causing progressive symptoms of memory loss and confusion. Although definite diagnosis is only possible by necropsy, differential diagnosis with other types of dementia is still needed. An electroencephalogram (EEG) is a cheap, portable, non-invasive method to record brain signals. Previous studies with non-linear signal processing methods have shown changes in the EEG due to AD, which is characterised reduced complexity and increased regularity. EEGs from 11 AD patients and 11 age-matched control subjects were analysed with Fuzzy Entropy (FuzzyEn), a non-linear method that was introduced as an improvement over the frequently used Approximate Entropy (ApEn) and Sample Entropy (SampEn) algorithms. AD patients had significantly lower FuzzyEn values than control subjects (p < 0.01) at electrodes T6, P3, P4, O1, and O2. Furthermore, when diagnostic accuracy was calculated using Receiver Operating Characteristic (ROC) curves, FuzzyEn outperformed both ApEn and SampEn, reaching a maximum accuracy of 86.36%. These results suggest that FuzzyEn could increase the insight into brain dysfunction in AD, providing potentially useful diagnostic information. However, results depend heavily on the input parameters that are used to compute FuzzyEn.

Complexity Changes in Brain Activity in Healthy Ageing: A Permutation Lempel-Ziv Complexity Study of Magnetoencephalograms.

Maturation and ageing, which can be characterised by the dynamic changes in brain morphology, can have an impact on the physiology of the brain. As such, it is possible that these changes can have an impact on the magnetic activity of the brain recorded using magnetoencephalography. In this study changes in the resting state brain (magnetic) activity due to healthy ageing were investigated by estimating the complexity of magnetoencephalogram (MEG) signals. The main aim of this study was to identify if the complexity of background MEG signals changed significantly across the human lifespan for both males and females. A sample of 177 healthy participants (79 males and 98 females aged between 21 and 80 and grouped into 3 categories i.e., early-, mid- and late-adulthood) was used in this investigation. This investigation also extended to evaluating if complexity values remained relatively stable during the 5 min recording. Complexity was estimated using permutation Lempel-Ziv complexity, a recently introduced complexity metric, with a motif length of 5 and a lag of 1. Effects of age and gender were investigated in the MEG channels over 5 brain regions, i.e., anterior, central, left lateral, posterior, and, right lateral, with highest complexity values observed in the signals recorded by the channels over the anterior and central regions of the brain. Results showed that while changes due to age had a significant effect on the complexity of the MEG signals recorded over 5 brain regions, gender did not have a significant effect on complexity values in all age groups investigated. Moreover, although some changes in complexity were observed between the different minutes of recording, due to the small magnitude of the changes it was concluded that practical significance might outweigh statistical significance in this instance. The results from this study can contribute to form a fingerprint of the characteristics of healthy ageing in MEGs that could be useful when investigating changes to the resting state activity due to pathology.

Lempel-Ziv complexity of cortical activity during sleep and waking in rats.

Understanding the dynamics of brain activity manifested in the EEG, local field potentials (LFP), and neuronal spiking is essential for explaining their underlying mechanisms and physiological significance. Much has been learned about sleep regulation using conventional EEG power spectrum, coherence, and period-amplitude analyses, which focus primarily on frequency and amplitude characteristics of the signals and on their spatio-temporal synchronicity. However, little is known about the effects of ongoing brain state or preceding sleep-wake history on the nonlinear dynamics of brain activity. Recent advances in developing novel mathematical approaches for investigating temporal structure of brain activity based on such measures, as Lempel-Ziv complexity (LZC) can provide insights that go beyond those obtained with conventional techniques of signal analysis. Here, we used extensive data sets obtained in spontaneously awake and sleeping adult male laboratory rats, as well as during and after sleep deprivation, to perform a detailed analysis of cortical LFP and neuronal activity with LZC approach. We found that activated brain states-waking and rapid eye movement (REM) sleep are characterized by higher LZC compared with non-rapid eye movement (NREM) sleep. Notably, LZC values derived from the LFP were especially low during early NREM sleep after sleep deprivation and toward the middle of individual NREM sleep episodes. We conclude that LZC is an important and yet largely unexplored measure with a high potential for investigating neurophysiological mechanisms of brain activity in health and disease.

Structural inequality and temporal brain dynamics across diverse samples.

BACKGROUND: Structural income inequality - the uneven income distribution across regions or countries - could affect brain structure and function, beyond individual differences. However, the impact of structural income inequality on the brain dynamics and the roles of demographics and cognition in these associations remains unexplored. METHODS: Here, we assessed the impact of structural income inequality, as measured by the Gini coefficient on multiple EEG metrics, while considering the subject-level effects of demographic (age, sex, education) and cognitive factors. Resting-state EEG signals were collected from a diverse sample (countries = 10; healthy individuals = 1394 from Argentina, Brazil, Colombia, Chile, Cuba, Greece, Ireland, Italy, Turkey and United Kingdom). Complexity (fractal dimension, permutation entropy, Wiener entropy, spectral structure variability), power spectral and aperiodic components (1/f slope, knee, offset), as well as graph-theoretic measures were analysed. FINDINGS: Despite variability in samples, data collection methods, and EEG acquisition parameters, structural inequality systematically predicted electrophysiological brain dynamics, proving to be a more crucial determinant of brain dynamics than individual-level factors. Complexity and aperiodic activity metrics captured better the effects of structural inequality on brain function. Following inequality, age and cognition emerged as the most influential predictors. The overall results provided convergent multimodal metrics of biologic embedding of structural income inequality characterised by less complex signals, increased random asynchronous neural activity, and reduced alpha and beta power, particularly over temporoposterior regions. CONCLUSION: These findings might challenge conventional neuroscience approaches that tend to overemphasise the influence of individual-level factors, while neglecting structural factors. Results pave the way for neuroscience-informed public policies aimed at tackling structural inequalities in diverse populations.

Brain clocks capture diversity and disparities in aging and dementia across geographically diverse populations.

Brain clocks, which quantify discrepancies between brain age and chronological age, hold promise for understanding brain health and disease. However, the impact of diversity (including geographical, socioeconomic, sociodemographic, sex and neurodegeneration) on the brain-age gap is unknown. We analyzed datasets from 5,306 participants across 15 countries (7 Latin American and Caribbean countries (LAC) and 8 non-LAC countries). Based on higher-order interactions, we developed a brain-age gap deep learning architecture for functional magnetic resonance imaging (2,953) and electroencephalography (2,353). The datasets comprised healthy controls and individuals with mild cognitive impairment, Alzheimer disease and behavioral variant frontotemporal dementia. LAC models evidenced older brain ages (functional magnetic resonance imaging: mean directional error = 5.60, root mean square error (r.m.s.e.) = 11.91; electroencephalography: mean directional error = 5.34, r.m.s.e. = 9.82) associated with frontoposterior networks compared with non-LAC models. Structural socioeconomic inequality, pollution and health disparities were influential predictors of increased brain-age gaps, especially in LAC (R² = 0.37, F² = 0.59, r.m.s.e. = 6.9). An ascending brain-age gap from healthy controls to mild cognitive impairment to Alzheimer disease was found. In LAC, we observed larger brain-age gaps in females in control and Alzheimer disease groups compared with the respective males. The results were not explained by variations in signal quality, demographics or acquisition methods. These findings provide a quantitative framework capturing the diversity of accelerated brain aging.

Brain clocks capture diversity and disparity in aging and dementia.

Brain clocks, which quantify discrepancies between brain age and chronological age, hold promise for understanding brain health and disease. However, the impact of multimodal diversity (geographical, socioeconomic, sociodemographic, sex, neurodegeneration) on the brain age gap (BAG) is unknown. Here, we analyzed datasets from 5,306 participants across 15 countries (7 Latin American countries -LAC, 8 non-LAC). Based on higher-order interactions in brain signals, we developed a BAG deep learning architecture for functional magnetic resonance imaging (fMRI=2,953) and electroencephalography (EEG=2,353). The datasets comprised healthy controls, and individuals with mild cognitive impairment, Alzheimer's disease, and behavioral variant frontotemporal dementia. LAC models evidenced older brain ages (fMRI: MDE=5.60, RMSE=11.91; EEG: MDE=5.34, RMSE=9.82) compared to non-LAC, associated with frontoposterior networks. Structural socioeconomic inequality and other disparity-related factors (pollution, health disparities) were influential predictors of increased brain age gaps, especially in LAC (R2=0.37, F2=0.59, RMSE=6.9). A gradient of increasing BAG from controls to mild cognitive impairment to Alzheimer's disease was found. In LAC, we observed larger BAGs in females in control and Alzheimer's disease groups compared to respective males. Results were not explained by variations in signal quality, demographics, or acquisition methods. Findings provide a quantitative framework capturing the multimodal diversity of accelerated brain aging.

Assessment and Comparison of Nonlinear Measures in Resting-State Magnetoencephalograms in Alzheimer's Disease and Mild Cognitive Impairment.

BACKGROUND: Nonlinear dynamical measures, such as fractal dimension (FD), entropy, and Lempel-Ziv complexity (LZC), have been extensively investigated individually for detecting information content in magnetoencephalograms (MEGs) from patients with Alzheimer's disease (AD). OBJECTIVE: To compare systematically the performance of twenty conventional and recently introduced nonlinear dynamical measures in studying AD versus mild cognitive impairment (MCI) and healthy control (HC) subjects using MEG. METHODS: We compared twenty nonlinear measures to distinguish MEG recordings from 36 AD (mean age = 74.06±6.95 years), 18 MCI (mean age = 74.89±5.57 years), and 26 HC subjects (mean age = 71.77±6.38 years) in different brain regions and also evaluated the effect of the length of MEG epochs on their performance. We also studied the correlation between these measures and cognitive performance based on the Mini-Mental State Examination (MMSE). RESULTS: The results obtained by LZC, zero-crossing rate (ZCR), FD, and dispersion entropy (DispEn) measures showed significant differences among the three groups. There was no significant difference between HC and MCI. The highest Hedge's g effect sizes for HC versus AD and MCI versus AD were respectively obtained by Higuchi's FD (HFD) and fuzzy DispEn (FuzDispEn) in the whole brain and was most prominent in left lateral. The results obtained by HFD and FuzDispEn had a significant correlation with the MMSE scores. DispEn-based techniques, LZC, and ZCR, compared with HFD, were less sensitive to epoch length in distinguishing HC form AD. CONCLUSIONS: FuzDispEn was the most consistent technique to distinguish MEG dynamical patterns in AD compared with HC and MCI.

Electroencephalogram Connectivity for the Diagnosis of Psychogenic Non-epileptic Seizures.

Psychogenic non-epileptic seizures (PNES) are attacks that resemble epilepsy but are not associated with epileptic brain activity and are regularly misdiagnosed. The current gold standard method of diagnosis is expensive and complex. Electroencephalogram (EEG) analysis with machine learning could improve this. A k-nearest neighbours (kNN) and support vector machine (SVM) were used to classify EEG connectivity measures from 48 patients with PNES and 29 patients with epilepsy. The synchronisation method - correlation or coherence - and the binarisation threshold were defined through experimentation. Ten network parameters were extracted from the synchronisation matrix. The broad, delta, theta, alpha, beta, gamma, and combined 'all' frequency bands were compared along with three feature selection methods: the full feature set (no selection), light gradient boosting machine (LGBM) and k-Best. Coherence was the highest performing synchronisation method and 0.6 was the best coherence threshold. The highest balanced accuracy was 89.74%, produced by combining all six frequency bands and selecting features with LGBM, classified by the SVM. This method returned a comparatively high accuracy but at a high computation cost. Future research should focus on identifying specific frequency bands and network parameters to reduce this cost. Clinical relevance - This study found that EEG connectivity and machine learning methods can be used to differentiate PNES from epilepsy using interictal recordings to a high accuracy. Thus, this method could be an effective tool in assisting clinicians in PNES diagnosis without a video- EEG recording of a habitual seizure.

Brain signal complexity in adults with Down syndrome: Potential application in the detection of mild cognitive impairment.

Background: Down syndrome (DS) is considered the most frequent cause of early-onset Alzheimer's disease (AD), and the typical pathophysiological signs are present in almost all individuals with DS by the age of 40. Despite of this evidence, the investigation on the pre-dementia stages in DS is scarce. In the present study we analyzed the complexity of brain oscillatory patterns and neuropsychological performance for the characterization of mild cognitive impairment (MCI) in DS. Materials and methods: Lempel-Ziv complexity (LZC) values from resting-state magnetoencephalography recordings and the neuropsychological performance in 28 patients with DS [control DS group (CN-DS) (n = 14), MCI group (MCI-DS) (n = 14)] and 14 individuals with typical neurodevelopment (CN-no-DS) were analyzed. Results: Lempel-Ziv complexity was lowest in the frontal region within the MCI-DS group, while the CN-DS group showed reduced values in parietal areas when compared with the CN-no-DS group. Also, the CN-no-DS group exhibited the expected pattern of significant increase of LZC as a function of age, while MCI-DS cases showed a decrease. The combination of reduced LZC values and a divergent trajectory of complexity evolution with age, allowed the discrimination of CN-DS vs. MCI-DS patients with a 92.9% of sensitivity and 85.7% of specificity. Finally, a pattern of mnestic and praxic impairment was significantly associated in MCI-DS cases with the significant reduction of LZC values in frontal and parietal regions (p = 0.01). Conclusion: Brain signal complexity measured with LZC is reduced in DS and its development with age is also disrupted. The combination of both features might assist in the detection of MCI within this population.

A novel deep learning approach using AlexNet for the classification of electroencephalograms in Alzheimer's Disease and Mild Cognitive Impairment.

Alzheimer's Disease (AD) is the most common form of dementia. Mild Cognitive Impairment (MCI) is the term given to the stage describing prodromal AD and represents a 'risk factor' in early-stage AD diagnosis from normal cognitive decline due to ageing. The electroencephalogram (EEG) has been studied extensively for AD characterization, but reliable early-stage diagnosis continues to present a challenge. The aim of this study was to introduce a novel way of classifying between AD patients, MCI subjects, and age-matched healthy control (HC) subjects using EEG-derived feature images and deep learning techniques. The EEG recordings of 141 age-matched subjects (52 AD, 37 MCI, 52 HC) were converted into 2D greyscale images representing the Pearson correlation coefficients and the distance Lempel-Ziv Complexity (dLZC) between the 21 EEG channels. Each feature type was computed from EEG epochs of 1s, 2s, 5s and 10s segmented from the original recording. The CNN architecture AlexNet was modified and employed for this three-way classification task and a 70/30 split was used for training and validation with each of the different epoch lengths and EEG-derived images. Whilst a maximum classification accuracy of 73.49% was obtained using dLZC-derived images from 10s epochs as input to the model, the classification accuracy reached 98.13% using the images obtained from Pearson correlation coefficients and 5s epochs. Clinical Relevance- The preliminary findings from this study show that deep learning applied to the analysis of the EEG can classify subjects with accuracies close to 100.

The correlation between white-matter microstructure and the complexity of spontaneous brain activity: a difussion tensor imaging-MEG study.

The advent of new signal processing methods, such as non-linear analysis techniques, represents a new perspective which adds further value to brain signals' analysis. Particularly, Lempel-Ziv's Complexity (LZC) has proven to be useful in exploring the complexity of the brain electromagnetic activity. However, an important problem is the lack of knowledge about the physiological determinants of these measures. Although a correlation between complexity and connectivity has been proposed, this hypothesis was never tested in vivo. Thus, the correlation between the microstructure of the anatomic connectivity and the functional complexity of the brain needs to be inspected. In this study we analyzed the correlation between LZC and fractional anisotropy (FA), a scalar quantity derived from diffusion tensors that is particularly useful as an estimate of the functional integrity of myelinated axonal fibers, in a group of sixteen healthy adults (all female, mean age 65.56±6.06 years, intervals 58-82). Our results showed a positive correlation between FA and LZC scores in regions including clusters in the splenium of the corpus callosum, cingulum, parahipocampal regions and the sagittal stratum. This study supports the notion of a positive correlation between the functional complexity of the brain and the microstructure of its anatomical connectivity. Our investigation proved that a combination of neuroanatomical and neurophysiological techniques may shed some light on the underlying physiological determinants of brain's oscillations.

Time course of cortical response complexity during extended wakefulness and its differential association with vigilance in young and older individuals.

Characterization of the complexity of electroencephalogram (EEG) responses has provided important insights in cognitive function as well as in the brain bases of consciousness and vigilance. Whether brain response complexity changes during prolonged wakefulness and sleep deprivation -when vigilance level considerably varies- is not fully elucidated yet. In the present study, we repeatedly assessed EEG responses to transcranial magnetic stimulation (TMS) over 34 h of sleep deprivation under constant routine conditions in healthy younger (N = 13; 5 women; 18-30 y) and older (N = 12; 6 women; 50-70 y) individuals, while they were performing a vigilance task. Response complexity was computed both at the global (all scalp sensors) and local (sensors surrounding TMS hotspot) levels using the Lempel-Ziv algorithm. Response complexity was significantly higher in the older compared to the young volunteers over the entire protocol. Global complexity response significantly changed with time spent awake, with an increasing trend from the beginning to the middle of the biological night, followed by a decreasing trend from the middle of the biological night to the following afternoon. An unexpected different link between vigilance performance and brain response complexity was detected across age groups: higher response complexity was associated with lower performance in the older group, particularly in the morning sessions. These findings show that cortical activity complexity changes with vigilance variation, as experienced during sleep deprivation and circadian misalignment, in two age groups, with no evident time course difference across age-groups. Aside from classical linear EEG analyses, computation of Lempel-Ziv complexity provides additional insights on the neurophysiology of the processes associated with vigilance and their modifications throughout ageing.

Deep learning of resting-state electroencephalogram signals for three-class classification of Alzheimer's disease, mild cognitive impairment and healthy ageing.

Objective.This study aimed to produce a novel deep learning (DL) model for the classification of subjects with Alzheimer's disease (AD), mild cognitive impairment (MCI) subjects and healthy ageing (HA) subjects using resting-state scalp electroencephalogram (EEG) signals.Approach.The raw EEG data were pre-processed to remove unwanted artefacts and sources of noise. The data were then processed with the continuous wavelet transform, using the Morse mother wavelet, to create time-frequency graphs with a wavelet coefficient scale range of 0-600. The graphs were combined into tiled topographical maps governed by the 10-20 system orientation for scalp electrodes. The application of this processing pipeline was used on a data set of resting-state EEG samples from age-matched groups of 52 AD subjects (82.3 ± 4.7 years of age), 37 MCI subjects (78.4 ± 5.1 years of age) and 52 HA subjects (79.6 ± 6.0 years of age). This resulted in the formation of a data set of 16197 topographical images. This image data set was then split into training, validation and test images and used as input to an AlexNet DL model. This model was comprised of five hidden convolutional layers and optimised for various parameters such as learning rate, learning rate schedule, optimiser, and batch size.Main results.The performance was assessed by a tenfold cross-validation strategy, which produced an average accuracy result of 98.9 ± 0.4% for the three-class classification of AD vs MCI vs HA. The results showed minimal overfitting and bias between classes, further indicating the strength of the model produced.Significance.These results provide significant improvement for this classification task compared to previous studies in this field and suggest that DL could contribute to the diagnosis of AD from EEG recordings.

Prediction of Paroxysmal Atrial Fibrillation From Complexity Analysis of the Sinus Rhythm ECG: A Retrospective Case/Control Pilot Study.

Paroxysmal atrial fibrillation (PAF) is the most common cardiac arrhythmia, conveying a stroke risk comparable to persistent AF. It poses a significant diagnostic challenge given its intermittency and potential brevity, and absence of symptoms in most patients. This pilot study introduces a novel biomarker for early PAF detection, based upon analysis of sinus rhythm ECG waveform complexity. Sinus rhythm ECG recordings were made from 52 patients with (n = 28) or without (n = 24) a subsequent diagnosis of PAF. Subjects used a handheld ECG monitor to record 28-second periods, twice-daily for at least 3 weeks. Two independent ECG complexity indices were calculated using a Lempel-Ziv algorithm: R-wave interval variability (beat detection, BD) and complexity of the entire ECG waveform (threshold crossing, TC). TC, but not BD, complexity scores were significantly greater in PAF patients, but TC complexity alone did not identify satisfactorily individual PAF cases. However, a composite complexity score (h-score) based on within-patient BD and TC variability scores was devised. The h-score allowed correct identification of PAF patients with 85% sensitivity and 83% specificity. This powerful but simple approach to identify PAF sufferers from analysis of brief periods of sinus-rhythm ECGs using hand-held monitors should enable easy and low-cost screening for PAF with the potential to reduce stroke occurrence.