Mutual Information-Based Brain Network Analysis in Post-stroke Patients With Different Levels of Depression

Functional Brain Network Disruptions in Parkinson's Disease: Insights from Information Theory and Machine Learning

Objectives: This study investigates disruptions in functional brain networks in Parkinson's Disease (PD), using advanced modeling and machine learning. Functional networks were constructed using the Nonlinear Autoregressive Distributed Lag (NARDL) model, which captures nonlinear and asymmetric dependencies between regions of interest (ROIs). Key network metrics and information-theoretic measures were extracted to classify PD patients and healthy controls (HC), using deep learning models, with explainability methods employed to identify influential features. Methods: Resting-state fMRI data from the Parkinson's Progression Markers Initiative (PPMI) dataset were used to construct NARDL-based networks. Metrics, such as Degree, Closeness, Betweenness, and Eigenvector Centrality, along with Network Entropy and Complexity, were analyzed. Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and Long Short-Term Memory (LSTM) models, classified PD and HC groups. Explainability techniques, including SHAP and LIME, identified significant features driving the classifications. Results: PD patients showed reduced Closeness (22%) and Betweenness Centrality (18%). CNN achieved 91% accuracy, with Network Entropy and Eigenvector Centrality identified as key features. Increased Network Entropy indicated heightened randomness in PD brain networks. Conclusions: NARDL-based analysis with interpretable deep learning effectively distinguishes PD from HC, offering insights into neural disruptions and potential personalized treatments for PD.

Similar connectivity of gut microbiota and brain activity networks is mediated by animal protein and lipid intake in children from a Mexican indigenous population

The gut microbiota-brain axis is a complex communication network essential for host health. Any long-term disruption can affect higher cognitive functions, or it may even result in several chronic neurological diseases. The type and diversity of nutrients an individual consumes are essential for developing the gut microbiota (GM) and the brain. Hence, dietary patterns might influence networks communication of this axis, especially at the age that both systems go through maturation processes. By implementing Mutual Information and Minimum Spanning Tree (MST); we proposed a novel combination of Machine Learning and Network Theory techniques to study the effect of animal protein and lipid intake on the connectivity of GM and brain cortex activity (BCA) networks in children from 5-to 10 years old from an indigenous community in the southwest of México. Socio-ecological conditions in this nonwestern lifestyle community are very homogeneous among its inhabitants but it shows high individual heterogeneity in the consumption of animal products. Results suggest that MST, the critical backbone of information flow, diminishes under low protein and lipid intake. So, under these nonwestern regimens, deficient animal protein and lipid consumption diets may significantly affect the GM-BCA connectivity in crucial development stages. Finally, MST offers us a metric that unifies biological systems of different nature to evaluate the change in their complexity in the face of environmental pressures or disturbances. Effect of Diet on gut microbiota and brain networks connectivity.

Reward network dysfunction is associated with cognitive impairment after stroke

Stroke survivors not only suffer from severe motor, speech and neurocognitive deficits, but in many cases also from a "lack of pleasure" and a reduced motivational level. Especially apathy and anhedonic symptoms can be linked to a dysfunction of the reward system. Rewards are considered as important co-factor for learning, so the question arises as to why and how this affects the rehabilitation of stroke patients. We investigated reward behaviour, learning ability and brain network connectivity in acute (3-7d) mild to moderate stroke patients (n = 28) and age-matched healthy controls (n = 26). Reward system activity was assessed using the Monetary Incentive Delay task (MID) during magnetoencephalography (MEG). Coherence analyses were used to demonstrate reward effects on brain functional network connectivity. The MID-task showed that stroke survivors had lower reward sensitivity and required greater monetary incentives to improve performance and showed deficits in learning improvement. MEG-analyses showed a reduced network connectivity in frontal and temporoparietal regions. All three effects (reduced reward sensitivity, reduced learning ability and altered cerebral connectivity) were found to be closely related and differed strongly from the healthy group. Our results reinforce the notion that acute stroke induces reward network dysfunction, leading to functional impairment of behavioural systems. These findings are representative of a general pattern in mild strokes and are independent of the specific lesion localisation. For stroke rehabilitation, these results represent an important point to identify the reduced learning capacity after stroke and to implement individualised recovery exercises accordingly.

Systems-level analysis of local field potentials reveals differential effects of lysergic acid diethylamide and ketamine on neuronal activity and functional connectivity

Psychedelic substances have in recent years attracted considerable interest as potential treatments for several psychiatric conditions, including depression, anxiety, and addiction. Imaging studies in humans point to a number of possible mechanisms underlying the acute effects of psychedelics, including changes in neuronal firing rates and excitability as well as alterations in functional connectivity between various brain nodes. In addition, animal studies using invasive recordings, have suggested synchronous high-frequency oscillations involving several brain regions as another key feature of the psychedelic brain state. To better understand how the imaging data might be related to high-resolution electrophysiological measurements, we have here analyzed the aperiodic part of the local field potential (LFP) in rodents treated with a classic psychedelic (LSD) or a dissociative anesthetic (ketamine). In addition, functional connectivity, as quantified by mutual information measures in the LFP time series, has been assessed with in and between different structures. Our data suggest that the altered brain states of LSD and ketamine are caused by different underlying mechanisms, where LFP power shifts indicate increased neuronal activity but reduced connectivity following ketamine, while LSD also leads to reduced connectivity but without an accompanying change in LFP broadband power.

Filtering ASVs/OTUs via mutual information-based microbiome network analysis

Microbial communities are widely studied using high-throughput sequencing techniques, such as 16S rRNA gene sequencing. These techniques have attracted biologists as they offer powerful tools to explore microbial communities and investigate their patterns of diversity in biological and biomedical samples at remarkable resolution. However, the accuracy of these methods can negatively affected by the presence of contamination. Several studies have recognized that contamination is a common problem in microbial studies and have offered promising computational and laboratory-based approaches to assess and remove contaminants. Here we propose a novel strategy, MI-based (mutual information based) filtering method, which uses information theoretic functionals and graph theory to identify and remove contaminants. We applied MI-based filtering method to a mock community data set and evaluated the amount of information loss due to filtering taxa. We also compared our method to commonly practice traditional filtering methods. In a mock community data set, MI-based filtering approach maintained the true bacteria in the community without significant loss of information. Our results indicate that MI-based filtering method effectively identifies and removes contaminants in microbial communities and hence it can be beneficial as a filtering method to microbiome studies. We believe our filtering method has two advantages over traditional filtering methods. First, it does not required an arbitrary choice of threshold and second, it is able to detect true taxa with low abundance.

Yijinjing Qigong intervention shows strong evidence on clinical effectiveness and electroencephalography signal features for early poststroke depression: A randomized, controlled trial

Objective

Although Traditional Chinese Yijinjing Qigong Exercise (YJJQE) as mind-body intervention is popularly used among adults to ameliorate depressive symptoms in China, no randomized controlled trials (RCTs) are available to evaluate the effects of YJJQE in patients with poststroke depression (PSD). This study aims to explore the clinical efficacy and the neurological and psychiatric mechanism in brain network functional connectivity underlying electroencephalography (EEG).

Materials and methods

A total of 60 patients, diagnosed with mild PSD, were randomly (1:1) assigned to YJJQE group (n = 30) and control group of routine segmental rehabilitation training group (n = 30) for a 60-min exercise session once a day for 3 weeks. All outcome measures were collected at baseline and 3-weeks ending intervention. The primary outcome was the 24-item Hamilton Depression Scale (HAMD-24) score, evaluation at more time points for 1 month of follow-up. The secondary outcomes were EEG data in four frequency domains (δ, θ, α, and β), global efficiency (GE), local efficiency (LE), GE/LE curve [areas under the curve (AUC)], Phase Lag Index (PLI), (HAMD-24) Score and EEG correlation analysis.

Results

All patients showed no significant differences in baseline data. After 3 weeks and 1 month of follow-up, the YJJQE group demonstrated significant decreasing changes compared to the control group on the HAMD-24 scores (p < 0.001). Furthermore, the YJJQE group also showed a significant reduction in θ wave, and an increase in both GE and LE. Compared to the control group, the YJJQE Qigong group showed significantly greater functional connectivity in the δ, θ, and β frequency bands in the brain network of the degree of phase synchronization (p < 0.001). HAMD-24 Score and EEG correlation analysis negative correlation in the Qigong group θ wave (p < 0.001).

Conclusion

Our findings demonstrated that YJJQE is estimated to effectively alleviate the depressed mood of patients with PSD by promoting the efficiency in information transmission of network functional connectivity and its integration ability in different brain regions. Therefore, the YJJQE would be useful as a non-pharmacological treatment to prevent PSD.

Clinical trial registration

[http://www.chictr.org.cn/showproj.aspx?proj=55789], identifier [ChiCTR2000035588].

Nonlinear functional muscle network based on information theory tracks sensorimotor integration post stroke

Sensory information is critical for motor coordination. However, understanding sensorimotor integration is complicated, especially in individuals with impairment due to injury to the central nervous system. This research presents a novel functional biomarker, based on a nonlinear network graph of muscle connectivity, called InfoMuNet, to quantify the role of sensory information on motor performance. Thirty-two individuals with post-stroke hemiparesis performed a grasp-and-lift task, while their muscle activity from 8 muscles in each arm was measured using surface electromyography. Subjects performed the task with their affected hand before and after sensory exposure to the task performed with the less-affected hand. For the first time, this work shows that InfoMuNet robustly quantifies changes in functional muscle connectivity in the affected hand after exposure to sensory information from the less-affected side. > 90% of the subjects conformed with the improvement resulting from this sensory exposure. InfoMuNet also shows high sensitivity to tactile, kinesthetic, and visual input alterations at the subject level, highlighting its potential use in precision rehabilitation interventions.

Reproducible Inter-Personal Brain Coupling Measurements in Hyperscanning Settings With functional Near Infra-Red Spectroscopy

Despite a huge advancement in neuroimaging techniques and growing importance of inter-personal brain research, few studies assess the most appropriate computational methods to measure brain-brain coupling. Here, we focus on the signal processing methods to detect brain-coupling in dyads. From a public dataset of functional Near Infra-Red Spectroscopy signals (N=24 dyads), we derived a synthetic control condition by randomization, we investigated the effectiveness of four most used signal similarity metrics: Cross Correlation, Mutual Information, Wavelet Coherence and Dynamic Time Warping. We also accounted for temporal variations between signals by allowing for misalignments up to a maximum lag. Starting from the observed effect sizes, computed in terms of Cohen's d, the power analysis indicated that a high sample size ([Formula: see text]) would be required to detect significant brain-coupling. We therefore discuss the need for specialized statistical approaches and propose bootstrap as an alternative method to avoid over-penalizing the results. In our settings, and based on bootstrap analyses, Cross Correlation and Dynamic Time Warping outperform Mutual Information and Wavelet Coherence for all considered maximum lags, with reproducible results. These results highlight the need to set specific guidelines as the high degree of customization of the signal processing procedures prevents the comparability between studies, their reproducibility and, ultimately, undermines the possibility of extracting new knowledge.

A novel nonlinear analysis of blood flow dynamics applied to the human lung

The spatial/temporal dynamics of blood flow in the human lung can be measured noninvasively with magnetic resonance imaging (MRI) using arterial spin labeling (ASL). We report a novel data analysis method using nonlinear prediction to identify dynamic interactions between blood flow units (image voxels), potentially providing a probe of underlying vascular control mechanisms. The approach first estimates the linear relationship (predictability) of one voxel time series with another using correlation analysis, and after removing the linear component estimates the nonlinear relationship with a numerical mutual information approach. Dimensionless global metrics for linear prediction (F_L) and nonlinear prediction (F_NL) represent the average amplitude of fluctuations in one voxel estimated by another voxel, as a percentage of the global average voxel flow. A proof-of-principle test of this approach analyzed experimental data from a study of high-altitude pulmonary edema (HAPE), providing two groups exhibiting known differences in vascular reactivity. Subjects were mountaineers divided into HAPE-susceptible (S, n=4) and HAPE-resistant (R, n=5) groups based on prior history at high altitude. Dynamic ASL measurements in the lung in normoxia (N, FIO₂=0.21) and hypoxia (H, FIO₂=0.13±0.01) were compared. The nonlinear prediction metric F_NL decreased with hypoxia (7.4±1.3(N) vs. 6.3±0.7(H), P=0.03) and was significantly different between groups (7.4±1.2 (R) vs. 6.2±14.1 (S), P=0.03). This proof-of-principle test demonstrates that this nonlinear analysis approach applied to ASL data is sensitive to physiological effects even in small subject cohorts, and potentially can be used in a wide range of studies in health and disease in the lung and other organs.

Neural Substrates of Poststroke Depression: Current Opinions and Methodology Trends

Poststroke depression (PSD), affecting about one-third of stroke survivors, exerts significant impact on patients’ functional outcome and mortality. Great efforts have been made since the 1970s to unravel the neuroanatomical substrate and the brain-behavior mechanism of PSD. Thanks to advances in neuroimaging and computational neuroscience in the past two decades, new techniques for uncovering the neural basis of symptoms or behavioral deficits caused by focal brain damage have been emerging. From the time of lesion analysis to the era of brain networks, our knowledge and understanding of the neural substrates for PSD are increasing. Pooled evidence from traditional lesion analysis, univariate or multivariate lesion-symptom mapping, regional structural and functional analyses, direct or indirect connectome analysis, and neuromodulation clinical trials for PSD, to some extent, echoes the frontal-limbic theory of depression. The neural substrates of PSD may be used for risk stratification and personalized therapeutic target identification in the future. In this review, we provide an update on the recent advances about the neural basis of PSD with the clinical implications and trends of methodology as the main features of interest.

A framework for quantifying the effects of transcranial magnetic stimulation on motor recovery from hemiparesis: Corticomuscular Network

OBJECTIVE

Transcranial magnetic stimulation (TMS) is an experimental therapy for promoting motor recovery from hemiparesis. At present, hemiparesis patients' responses to TMS are variable. To maximize its therapeutic potential, we need an approach that relates the electrophysiology of motor recovery and TMS. To this end, we propose Corticomuscular Network (CMN) representing the holistic motor system, including the cortico-cortical pathway, corticospinal tract, and muscle co-activation.

METHODS

CMN is made up of coherence between pairs of electrode signals and spatial locations of the electrodes. We associated coherence and graph features of CMN with Fugl-Meyer Assessment (FMA) for the upper extremity. Besides, we compared CMN between 8 patients with hemiparesis and 6 healthy controls and contrasted CMN of patients before and after a 1Hz TMS.

MAIN RESULTS

Corticomuscular coherence (CMC) correlated positively with FMA. The regression model between FMA and CMC between 5 pairs of channels had 0.99 adjusted R^2 and a p-value less than 0.01. Compared to healthy controls, CMN of patients tended to be a small-world network and was more interconnected with higher CMC. CMC between cortex and triceps brachii long head was higher in patients. 15-minute 1Hz TMS protocol induced coherence changes beyond the stimulation side and had a limited impact on CMN parameters that are related to motor recovery.

SIGNIFICANCE

CMN is a potential clinical approach to quantify rehabilitating progress. It also sheds light on the desirable electrophysiological effects of TMS based on which rehabilitating strategies can be optimized.

EEG functional connectivity is sensitive for nitrogen narcosis at 608 kPa

Divers commonly breathe air, containing nitrogen. Nitrogen under hyperbaric conditions is a narcotic gas. In dives beyond a notional threshold of 30 m depth (405 kPa) this can cause cognitive impairment, culminating in accidents due to poor decision making. Helium is known to have no narcotic effect. This study explored potential approaches to developing an electroencephalogram (EEG) functional connectivity metric to measure narcosis produced by nitrogen at hyperbaric pressures. Twelve human participants (five female) breathed air and heliox (in random order) at 284 and 608 kPa while recording 32-channel EEG and psychometric function. The degree of spatial functional connectivity, estimated using mutual information, was summarized with global efficiency. Air-breathing at 608 kPa (experienced as mild narcosis) caused a 35% increase in global efficiency compared to surface air-breathing (mean increase = 0.17, 95% CI [0.09–0.25], p = 0.001). Air-breathing at 284 kPa trended in a similar direction. Functional connectivity was modestly associated with psychometric impairment (mixed-effects model r² = 0.60, receiver-operating-characteristic area, 0.67 [0.51–0.84], p = 0.02). Heliox breathing did not cause a significant change in functional connectivity. In conclusion, functional connectivity increased during hyperbaric air-breathing in a dose-dependent manner, but not while heliox-breathing. This suggests sensitivity to nitrogen narcosis specifically.

The Structural and Functional Neuroanatomy of Post-Stroke Depression and Executive Dysfunction: A Review of Neuroimaging Findings and Implications for Treatment

Post-stroke depression and executive dysfunction co-occur and are highly debilitating. Few treatments alleviate both depression and executive dysfunction after stroke. Understanding the brain network changes underlying post-stroke depression with executive dysfunction can inform the development of targeted and efficacious treatment. In this review, we synthesize neuroimaging findings in post-stroke depression and post-stroke executive dysfunction and highlight the network commonalities that may underlie this comorbidity. Structural and functional alterations in the cognitive control network, salience network, and default mode network are associated with depression and executive dysfunction after stroke. Specifically, post-stroke depression and executive dysfunction are both linked to changes in intrinsic functional connectivity within resting state networks, functional over-connectivity between the default mode and salience/cognitive control networks, and reduced cross-hemispheric frontoparietal functional connectivity. Cognitive training and noninvasive brain stimulation targeted at these brain network abnormalities and specific clinical phenotypes may help advance treatment for post-stroke depression with executive dysfunction.

Cortical network modularity changes along the course of frontotemporal and Alzheimer's dementing diseases

Cortical network modularity underpins cognitive functions, so we hypothesized its progressive derangement along the course of frontotemporal (FTD) and Alzheimer's (AD) dementing diseases. EEG was recorded in 18 FTD, 18 AD, and 20 healthy controls (HC). In the FTD and AD patients, the EEG recordings were performed at the prodromal stage of dementia, at the onset of dementia, and three years after the onset of dementia. HC underwent three EEG recordings at 2-3-year time interval. Information flows underlying EEG activity recorded at electrode pairs were estimated by means of Mutual Information (MI) analysis. The functional organization of the cortical network was modelled by means of the Graph theory analysis on MI adjacency matrices. Graph theory analysis showed that the main hub of HC (Parietal area) was lost in FTD patients at onset of dementia, substituted by provincial hubs in frontal leads. No changes in global network organization were found in AD. Despite a progressive cognitive impairment during the FTD and AD progression, only the FTD patients showed a derangement in the cortical network modularity, possibly due to dysfunctions in frontal functional connectivity.

Analysis of Functional Brain Network in MDD Based on Improved Empirical Mode Decomposition With Resting State EEG Data

At present, most brain functional studies are based on traditional frequency bands to explore the abnormal functional connections and topological organization of patients with depression. However, they ignore the characteristic relationship of electroencephalogram (EEG) signals in the time domain. Therefore, this paper proposes a network decomposition model based on Improved Empirical Mode Decomposition (EMD), it is suitable for time-frequency analysis of brain functional network. On the one hand, it solves the problem of mode mixing on original EMD method, especially on high-density EEG data. On the other hand, by building brain function networks on different intrinsic mode function (IMF), we can perform time-frequency analysis of brain function connections. It provides a new insight for brain function connectivity analysis of major depressive disorder (MDD). Experimental results found that the IMFs waveform decomposed by Improved EMD was more stable and the difference between IMFs was obvious, it indicated that the mode mixing can be effectively solved. Besides, the analysis of the brain network, we found that the changes in MDD functional connectivity on different IMFs, it may be related to the pathological changes for MDD. More statistical results on three network metrics proved that there were significant differences between MDD and normal controls (NC) group. In addition, the aberrant brain network structure of MDDs was also confirmed in the hubs characteristic. These findings may provide potential biomarkers for the clinical diagnosis of MDD patients.

Characterization of kinesthetic motor imagery compared with visual motor imageries

Motor imagery (MI) is the only way for disabled subjects to robustly use a robot arm with a brain-machine interface. There are two main types of MI. Kinesthetic motor imagery (KMI) is proprioceptive (OR somato-) sensory imagination and Visual motor imagery (VMI) represents a visualization of the corresponding movement incorporating the visual network. Because these imagery tactics may use different networks, we hypothesized that the connectivity measures could characterize the two imageries better than the local activity. Electroencephalography data were recorded. Subjects performed different conditions, including motor execution (ME), KMI, VMI, and visual observation (VO). We tried to classify the KMI and VMI by conventional power analysis and by the connectivity measures. The mean accuracies of the classification of the KMI and VMI were 98.5% and 99.29% by connectivity measures (alpha and beta, respectively), which were higher than those by the normalized power (p < 0.01, Wilcoxon paired rank test). Additionally, the connectivity patterns were correlated between the ME-KMI and between the VO-VMI. The degree centrality (DC) was significantly higher in the left-S1 at the alpha-band in the KMI than in the VMI. The MI could be well classified because the KMI recruits a similar network to the ME. These findings could contribute to MI training methods.

Functional Connectivity Analysis on Resting-State Electroencephalography Signals Following Chiropractic Spinal Manipulation in Stroke Patients

Stroke impairments often present as cognitive and motor deficits, leading to a decline in quality of life. Recovery strategy and mechanisms, such as neuroplasticity, are important factors, as these can help improve the effectiveness of rehabilitation. The present study investigated chiropractic spinal manipulation (SM) and its effects on resting-state functional connectivity in 24 subacute to chronic stroke patients monitored by electroencephalography (EEG). Functional connectivity of both linear and non-linear coupling was estimated by coherence and phase lag index (PLI), respectively. Non-parametric cluster-based permutation tests were used to assess the statistical significance of the changes in functional connectivity following SM. Results showed a significant increase in functional connectivity from the PLI metric in the alpha band within the default mode network (DMN). The functional connectivity between the posterior cingulate cortex and parahippocampal regions increased following SM, t (23) = 10.45, p = 0.005. No significant changes occurred following the sham control procedure. These findings suggest that SM may alter functional connectivity in the brain of stroke patients and highlights the potential of EEG for monitoring neuroplastic changes following SM. Furthermore, the altered connectivity was observed between areas which may be affected by factors such as decreased pain perception, episodic memory, navigation, and space representation in the brain. However, these factors were not directly monitored in this study. Therefore, further research is needed to elucidate the underlying mechanisms and clinical significance of the observed changes.