Most recent paper

Commun Biol. 2025 Aug 27;8(1):1287. doi: 10.1038/s42003-025-08703-7.

ABSTRACT

While traditionally regarded as "noise", blood-oxygenation-level-dependent (BOLD) functional Magnetic Resonance Imaging (fMRI) fluctuations coupled to systemic physiology-such as heart rate and respiratory changes-also hold valuable information about brain vascular properties and autonomic function. In this study, we leverage these physiological signals to characterize age-related changes in brain physiology, drawing on a large dataset from the Lifespan Human Connectome Project Aging study. Our findings reveal that aging is associated with globally slower respiratory fMRI responses, alongside faster cardiac fMRI responses and enhanced brain-cardiac signal coupling. Moreover, we show that the impact of age on physiological fMRI signals exhibits a notable turning point after age 60, suggesting a critical role of declining vascular health and autonomic function in aging. The potential impact of age-related changes in brain structure, tissue perfusion, and in-scan arousal states on the identified physiological fMRI patterns is also tested and discussed. Altogether, our results underscore significant age effects in the fMRI signatures of systemic physiology, emphasizing the pivotal role of altered vascular properties and autonomic function in aging. Methodologically, this study also demonstrates the utility of resting-state fMRI for extracting multi-parametric information about brain physiology, offering new biomarker opportunities that complement established functional connectivity metrics.

PMID:40858848 | DOI:10.1038/s42003-025-08703-7

Eur J Phys Rehabil Med. 2025 Jun;61(3):462-471. doi: 10.23736/S1973-9087.25.08872-0.

ABSTRACT

BACKGROUND: Visuospatial function is a core domain of functional cognition in stroke. Post-stroke cognitive impairment disrupts rehabilitation practice, highlighting the importance of characterizing patients with higher-order visuospatial dysfunction to inform personalized rehabilitation strategies. Although neuroimaging offers insights into disease-related mechanisms, its clinical application remains limited.

AIM: The aim of this paper was to investigate whether the residual resting-state functional connectivity supports higher-order visuospatial function after stroke and whether changes in connectivity can characterize patients with visuospatial dysfunction.

DESIGN: Observational study.

SETTING: Inpatient rehabilitation ward at Fujita Health University Hospital in Japan.

POPULATION: Fifty-eight patients with stroke.

METHODS: Visuospatial analogical reasoning was assessed using Raven's Colored Progressive Matrices (RCPM). Resting-state functional connectivity was evaluated using functional magnetic resonance imaging (fMRI). Empirical covariance matrices and group-sparse inverse covariance (GSIC) matrices were computed from the fMRI data, with the latter negated to estimate partial correlations representing direct connectivity. Correlations between connectivity measures and RCPM scores were analyzed, alongside data-driven clustering to stratify patients.

RESULTS: No significant correlation was found between empirical covariance connectivity and RCPM scores. However, GSIC-based analysis revealed a significant inverse correlation between connectivity of the posteromedial and the left inferior parietal cortex and RCPM scores. Higher parietal connectivity was associated with lower RCPM performance. Patients in the highest connectivity cluster exhibited severe impairments in visuospatial analogical reasoning, particularly in tasks requiring the integration of discrete figures into spatially related wholes. The lesions in these patients were predominantly localized in the left subcortex.

CONCLUSIONS: Medio-lateral parietal connectivity may underlie visuospatial analogical reasoning after stroke.

CLINICAL REHABILITATION IMPACT: Clustering analysis highlighted a distinct pattern of low scores in patients with increased parietal connectivity, suggesting that parietal connectivity changes have the potential for characterizing patients with severe dysfunction.

PMID:40856377 | DOI:10.23736/S1973-9087.25.08872-0

Psychol Med. 2025 Aug 26;55:e245. doi: 10.1017/S0033291725101451.

ABSTRACT

BACKGROUND: Methadone maintenance treatment (MMT) and protracted abstinence (PA) effectively reduce the craving for heroin among individuals with heroin use disorder (HUD). However, the difference in their effects on brain function, especially the coupling among the large-scale brain networks (default mode [DMN], salience [SN], and executive control [ECN] networks), remains unclear. This study analyzed the effects of the MMT and PA on these networks and the predictive value of the bilateral resource allocation index (RAI) for craving for heroin.

METHODS: Twenty-five individuals undergoing the MMT, 22 undergoing the PA, and 51 healthy controls underwent resting-state functional magnetic resonance imaging (rs-fMRI). Independent component analysis identified the ECN, DMN, and SN. The SN-ECN and SN-DMN connectivity and the bilateral RAI were evaluated. The relationships between network coupling and clinical and psychological characteristics were analyzed. The multiple linear regression model identified significant variables for predicting craving scores.

RESULTS: The MMT group showed significantly stronger SN-left ECN (lECN) coupling and left RAI than the PA group. In the MMT group, SN-lECN connectivity and bilateral RAI were positively correlated with the total methadone dose. In both treatment groups, SN-right ECN (rECN) connectivity and right RAI were negatively correlated with craving. The models revealed that the bilateral RAI and the MMT and PA were associated with the craving.

CONCLUSIONS: The MMT enhances SN-lECN coupling and the left RAI more than the PA, possibly due to higher control modulation. The RAI could help predict heroin craving in individuals with HUD undergoing either treatment program.

PMID:40856291 | DOI:10.1017/S0033291725101451

Proc IEEE Int Symp Biomed Imaging. 2024 May;2024. doi: 10.1109/isbi56570.2024.10635270. Epub 2024 Aug 22.

ABSTRACT

High-resolution fMRI provides a window into the brain's mesoscale organization. Yet, higher spatial resolution increases scan times, to compensate for the low signal and contrast-to-noise ratio. This work introduces a deep learning-based 3D super-resolution (SR) method for fMRI. By incorporating a resolution-agnostic image augmentation framework, our method adapts to varying voxel sizes without retraining. We apply this innovative technique to localize fine-scale motion-selective sites in the early visual areas. Detection of these sites typically requires ≤ 1mm isotropic data, whereas here, we visualize them based on lower resolution (2-3mm isotropic) fMRI data. Remarkably, the super-resolved fMRI is able to recover high-frequency detail of the interdigitated organization of these sites (relative to the color-selective sites), even with training data sourced from different subjects and experimental paradigms - including non-visual resting-state fMRI, underscoring its robustness and versatility. Quantitative and qualitative results indicate that our method has the potential to enhance the spatial resolution of fMRI, leading to a drastic reduction in acquisition time.

PMID:40855854 | PMC:PMC12376370 | DOI:10.1109/isbi56570.2024.10635270

Psychiatry Clin Neurosci. 2025 Aug 25. doi: 10.1111/pcn.13887. Online ahead of print.

ABSTRACT

AIM: Cognition varies across people with psychosis, including within a specific diagnosis. An important issue is identifying psychosis-specific neuro-cognitive dysfunctions. We addressed this issue by studying patterns of relationships between cognition and multiple other measures in persons with psychosis, their first-degree biological relatives, and healthy individuals (largest possible n = 2826).

METHODS: Brief Assessment of Cognition and Wide Range Achievement Test estimated cognitive performance. Neuroanatomical measures were FreeSurfer parcellations of 3T MRI structural brain scans. Brain functioning measures included saccades, smooth pursuit eye movements, stop signal, EEG, ERPs, resting state fMRI, plus clinical characteristics. Overall associations between 452 measures of brain structure-function and clinical characteristics (predictors) with cognitive performance (criterion) were estimated using the High Dimensional Empirical Bayes Screening algorithm.

RESULTS: The model yielded a common slope of predictors on cognitive performance (slope = 0.18, r = 0.33, P < 0.001). The majority (85%) of predictors fit this function, called the BAsic NeuroCognitive Continuum (BANCC). This relationship was stronger for psychosis probands (slope = 0.20, r = 0.38) than for relatives (slope = 0.09, r = 0.17) and healthy persons (slope = 0.11, r = 0.22). There were predictor-specific deviations from the common slope. Variables more strongly associated with cognitive performance (frontal-temporal-parietal lobe volumes, hippocampal regions, antisaccade performance) may tap neural architecture common to primary psychosis pathology. Variables unrelated to cognitive performance (intrinsic neural activity, volumes of lateral thalamic nuclei) distinguish specific neurophysiologically defined B-SNIP psychosis Biotypes and may capture signatures of psychosis pathophysiology.

DISCUSSION: BANCC is identifiable across humans, but deviations from that common attribute identify features of brain structure-function perhaps most central and specific to psychosis-related pathophysiology.

PMID:40855769 | DOI:10.1111/pcn.13887

Clin Neuroimaging (Hoboken). 2025;2(1):e70022. doi: 10.1002/neo2.70022. Epub 2025 Jun 26.

ABSTRACT

BACKGROUND AND PURPOSE: Neurogenic lower urinary tract dysfunction (NLUTD) affects over 80% of individuals with multiple sclerosis (MS), leading to significant morbidity and mortality due to storage and voiding dysfunction. This study aims to investigate the altered functional connectivity (FC) in cerebellar regions involved in bladder control in women with MS and NLUTD, compared to healthy controls, in both empty and full bladder states using concurrent urodynamics and functional magnetic resonance imaging (fMRI).

METHODS: We recruited 11 women with clinically stable MS and NLUTD and 10 healthy controls. Brain imaging data was collected using 7T MRI scanners, and functional connectivity was analyzed with three cerebellar regions of interest (ROIs) associated with bladder control. Functional connectivity data was processed using the CONN toolbox, and FC patterns were compared between groups during both resting empty and full bladder states.

RESULTS: In the empty bladder state, MS patients exhibited stronger intracerebellar FC, particularly in the right Crus 1, suggesting decreased motor control of the pelvic floor. Additionally, decreased FC was observed in the precuneus and prefrontal cortex, regions associated with bladder control. During the full bladder state, MS patients showed decreased FC in temporal, occipital, and prefrontal cortex, indicating impaired executive control over voiding.

CONCLUSION: This study highlights altered cerebellar connectivity in MS patients with NLUTD, providing novel insights into the neural mechanisms underlying bladder dysfunction and identifying potential therapeutic targets for restoring continence.

PMID:40852051 | PMC:PMC12369983 | DOI:10.1002/neo2.70022

Imaging Neurosci (Camb). 2025 Aug 20;3:IMAG.a.114. doi: 10.1162/IMAG.a.114. eCollection 2025.

ABSTRACT

Independent component analysis (ICA) denoising methods can be highly effective for reducing functional magnetic resonance imaging (fMRI) noise. ICA denoising method success heavily depends, however, on the accurate classification of fMRI data ICs as either neural signal or noise. While manual IC classification ("manual ICA denoising") is a current gold-standard, it requires extensive time and training. Automated methods of IC classification ("automated ICA denoising"), meanwhile, are less accurate and effective, especially in clinical populations where motion artifacts are more common. To address these challenges, a novel denoising method, Comprehensive Independent Component Analysis Denoising Assistant (CICADA), was developed. Uniquely, CICADA uses manual classification guidelines to automatically, comprehensively, and accurately capture most common sources of fMRI noise. As such, we hypothesized that CICADA would perform similarly to manual ICA denoising and outperform other current automated denoising methods. CICADA was evaluated against two well-established automated ICA denoising methods (FIX and ICA-AROMA) across three fMRI datasets. The datasets included high-motion resting-state (N = 57) and visual-task data (N = 53), both from individuals with schizophrenia, as well as low-motion resting-state healthy control data from an openly available dataset (N = 56). IC classification accuracy was first evaluated against manual IC classification in a subset (N = 30) of each dataset. Denoising performance efficacy was then evaluated with commonly used quality control (QC) benchmarks and correlations with fMRI noise profiles across all data. With a 97.9% mean overall accuracy in IC classification, CICADA performed nearly as well as manual IC classification and was significantly more accurate than FIX (92.9% mean overall accuracy; all p-values < 0.01) and ICA-AROMA (83.8% mean overall accuracy; all p-values < 0.001). CICADA also matched or outperformed FIX and ICA-AROMA across most QC and noise profile metrics across all data. Furthermore, CICADA greatly eased implementation of manual ICA denoising by decreasing the number of ICs a user must inspect by an average of 75%. Overall, CICADA is a novel, accurate, comprehensive, and automated ICA denoising tool for use in both resting-state and task-based fMRI. It performed similarly to the labor-intensive manual IC classification gold-standard and, in some datasets, outperformed current automated ICA denoising methods. Finally, CICADA may facilitate more efficient manual ICA denoising without reducing efficacy.

PMID:40851911 | PMC:PMC12368612 | DOI:10.1162/IMAG.a.114

J Affect Disord. 2025 Aug 22:120106. doi: 10.1016/j.jad.2025.120106. Online ahead of print.

ABSTRACT

BACKGROUND: Altered functional connectivity dynamics are observed in major depressive disorder (MDD), anxiety disorders (ANX), and childhood trauma (CT), but their combined impact remains unclear. Given their frequent co-occurrence and potential shared neural mechanisms, this study examines connectivity state dynamics and spatial organization in individuals with MDD and/or ANX, with an additional focus on CT.

METHODS: Resting-state fMRI data were acquired from 150 individuals with MDD and/or ANX (N = 86 with CT, N = 64 without CT) and 57 non-affected controls. Functional connectivity states were identified using k-means clustering on edge time-series. State dynamics (total transitions, fractional occupancy) and topological characteristics (mean connectivity, eigenvector centrality, modularity, global efficiency) were compared between groups (MDD/ANX vs. controls; MDD/ANX + CT vs. MDD/ANX-noCT vs. controls).

RESULTS: Four connectivity states were identified. No group differences in state dynamics were observed. However, MDD/ANX individuals exhibited altered topology in the weakly connected state (state 3), with reduced modularity and increased global efficiency compared to controls. In the sensory processing state (state 2), MDD/ANX + CT individuals showed lower SMN centrality compared to controls but not MDD/ANX-noCT, although this effect disappeared in a sensitivity analysis excluding controls with CT. No significant associations were found with depression, anxiety, or CT severity.

CONCLUSION: Altered topological organization reflecting less functional segregation in specific connectivity states was observed in MDD/ANX, suggesting impaired information processing. No distinct differences emerged between clinical groups with and without CT, suggesting that the observed alterations primarily reflect effects of MDD/ANX rather than CT status.

PMID:40850548 | DOI:10.1016/j.jad.2025.120106

Hear Res. 2025 Aug 18;466:109407. doi: 10.1016/j.heares.2025.109407. Online ahead of print.

ABSTRACT

PURPOSE: To explore functional connectivity alterations in idiopathic tinnitus patients at the integrity, network, and edge levels through resting-state functional magnetic resonance imaging (rs-fMRI), and elucidating central plasticity changes in idiopathic tinnitus from the perspective of central functional connectivity patterns.

METHOD: Collect rs-fMRI data from 74 patients with idiopathic tinnitus and 98 healthy volunteers, analyze the functional connectivity differences between the two groups at the integrity, network, and edge levels, and explore the correlation between these differences and the clinical characteristics of idiopathic tinnitus patients.

RESULT: Patients with idiopathic tinnitus significantly reduce node degree in the left parahippocampal gyrus, left amygdala, and bilateral Heschl gyrus. At the network level, the intranetwork functional connectivity strength in the auditory network (AUN) of idiopathic tinnitus patients is significantly reduced, and the internetwork connectivity strength between the auditory network (AUN) and the ventral attention network (VAN) is significantly reduced. In addition, we found that the functional connectivity strength of two pairs of regions of interest (ROI) increased dramatically at the edge level, while the functional connectivity strength of nine pairs of ROI significantly decreased.

CONCLUSION: Central functional connectivity reorganization in idiopathic tinnitus patients involves not only the auditory cortex but also the limbic system and cortical regions responsible for attention shifting.

PMID:40850190 | DOI:10.1016/j.heares.2025.109407

Neuroimage Clin. 2025 Aug 22:103869. doi: 10.1016/j.nicl.2025.103869. Online ahead of print.

NO ABSTRACT

PMID:40849253 | DOI:10.1016/j.nicl.2025.103869

J Neuroimaging. 2025 Jul-Aug;35(4):e70075. doi: 10.1111/jon.70075.

ABSTRACT

BACKGROUND AND PURPOSE: Resting-state fMRI (rs-fMRI) is increasingly used to map brain networks in patients under anesthesia, but technical factors can affect its utility. We evaluated the effects of sevoflurane, multiband acceleration, and scan duration on rs-fMRI signal quality and within-subject reliability under anesthesia.

METHODS: We retrospectively analyzed 64 clinical rs-fMRI scans acquired under anesthesia, with or without sevoflurane and multiband factor 5 acceleration. Temporal signal-to-noise ratio (tSNR) was used as a measure of signal quality. For each patient, the scan was split in half, and seed-based connectivity maps were generated for the primary motor cortex (M1), posterior cingulate cortex (PCC), and subgenual anterior cingulate cortex (sgACC). Split-half spatial correlations were used to assess within-subject reliability. Group comparisons examined differences in tSNR and reliability across conditions, and correlations with scan duration were tested.

RESULTS: Multiband acceleration was associated with significantly lower tSNR (U = 652.0, p = 8.9 × 10─6) and reduced split-half reliability for M1 (p = 0.019), PCC (p = 0.010), and sgACC (p = 0.0064). Sevoflurane showed no significant effect on tSNR or reliability. Longer scan duration correlated with improved reliability for M1 (r = 0.38, p = 0.003) but not for PCC or sgACC. No correlation was found between tSNR and reliability.

CONCLUSION: Hight multiband acceleration reduces both signal quality and reliability of rs-fMRI under anesthesia. Sevoflurane had no measurable effect. The lack of correlation between tSNR and reliability underscores the need for more robust metrics when evaluating scan quality.

PMID:40848013 | DOI:10.1111/jon.70075

Brain Behav. 2025 Aug;15(8):e70751. doi: 10.1002/brb3.70751.

ABSTRACT

OBJECTIVE: The perception of pain is thought to arise from the integration of information between multiple brain regions. Data from observational studies indicates that dysfunction of brain resting-state functional networks is present in a wide range of peripheral neuropathic pain (pNP). The present study thus sought to investigate whether a causal relationship exists and to determine the potential mediating role of circulating inflammatory proteins in this association.

METHODS: The resting-state functional magnetic imaging phenotype is defined as a stable feature that quantifies the pattern of functional connectivity (i.e., synchronized activity) between different regions of the brain in the resting state of an individual. We gathered publicly available genome-wide association study (GWAS) summary statistics for brain functional networks, including 191 rsfMRI phenotypes and postherpetic neuralgia (PHN) and trigeminal neuralgia (TN) in the FinnGen biobank. Furthermore, data were collected on genetic variation related to inflammation, including 91 circulating inflammatory proteins. We performed two-sample MR analysis to investigate the causal effects of functional brain networks on PHN and TN. To explore the possible mediation of inflammatory factor changes between rsfMRI phenotypes and PHN and TN.

RESULTS: The forward MR approach identifies five rs-fMRI phenotypes that are causally associated with the risk of developing PHN. For instance, enhanced motor network connectivity was found to be associated with a reduced risk of PHN. Six rsfMRI phenotypes were identified as causally associated with TN risk. These brain network phenotypes mainly involve the default mode network (DMN), the sensory-motor network (SMN), and the motor network, etc. Two-step MR-mediated analysis revealed that the inflammatory protein interleukin 20 receptor alpha (IL-20RA) is a mediator of the pathway from the phenotype Pheno 12 of the brain motor network to PHN.

CONCLUSION: The findings provide valuable insights into potential targets for disease intervention and treatment at the level of functional brain networks.

PMID:40847468 | PMC:PMC12373714 | DOI:10.1002/brb3.70751

Hear Res. 2025 Aug 18;466:109406. doi: 10.1016/j.heares.2025.109406. Online ahead of print.

ABSTRACT

Tinnitus, the perception of sound without an external source, is increasingly recognized as a disorder of large-scale brain networks rather than isolated auditory dysfunction. Aging introduces widespread neurobiological changes including cortical atrophy, reduced inhibitory control, and altered connectivity that may distinctly shape the neural basis of tinnitus across the adult lifespan. This study investigates age-specific brain network alterations in chronic subjective tinnitus using resting-state functional magnetic resonance imaging (rs-fMRI) and diffusion tensor imaging (DTI). Sixty tinnitus patients and sixty age-matched controls were stratified into younger (19-35 years) and older (45-65 years) cohorts. We assessed spontaneous activity via amplitude of low-frequency fluctuations (ALFF), local functional connectivity using local correlation (LCOR), and structural-functional coupling through voxel-wise correlations between ALFF and fractional anisotropy (FA). Younger tinnitus patients exhibited hyperactivity in auditory and limbic regions, alongside reduced prefrontal and occipital regulation, suggesting heightened sensory salience and impaired habituation. In contrast, older patients showed increased activation in frontal and cerebellar regions but diminished integration in default mode and attention-related cortices, indicating compensatory recruitment constrained by age-related decline. Structure-function coupling revealed more diffuse and adaptive correlations in younger patients, while older adults exhibited focal and limited coupling, particularly in temporal and occipital regions. Network alterations were differentially modulated by tinnitus severity and anxiety across age groups. These findings reveal distinct, age-contingent neural signatures of tinnitus, emphasizing the role of aging in modulating brain network plasticity and underscoring the need for lifespan-informed models in tinnitus research.

PMID:40845546 | DOI:10.1016/j.heares.2025.109406

Biol Psychiatry Cogn Neurosci Neuroimaging. 2025 Aug 19:S2451-9022(25)00250-2. doi: 10.1016/j.bpsc.2025.08.002. Online ahead of print.

ABSTRACT

BACKGROUND: Posttraumatic stress disorder (PTSD) and mild traumatic brain injury (mTBI) are associated with alterations in the functional connectome, specifically in canonical resting state networks including the default mode (DMN), central executive (CEN), and salience networks (SN). Comorbid PTSD+mTBI is linked to worse functional outcomes, but little is known about effects on the functional connectome.

METHODS: We investigated brain phenotypes from resting-state fMRI associated with PTSD (n=326), mTBI (n=448), and comorbid PTSD+mTBI (n=289) in military veterans and civilians (n=1526) from ENIGMA-TBI and -PTSD. We examined static functional connectivity (SFC) and dynamic functional connectivity (DFC), quantified both as variability in FC (VFC) over time and as dwell time in recurring FC states identified through clustering. ANCOVA was followed by post-hoc linear regression to test main and interaction effects of diagnosis on FC metrics.

RESULTS: We found a significant (pFDR<0.05) interaction of diagnosis by age on VFC. Older comorbid subjects had greater VFC within SN, between SN-to-CEN and SN-to-DMN than older controls. Comorbid relative to control subjects had significantly greater dwell time in an externally focused state. Comorbid and mTBI groups, relative to control subjects, had greater dwell time in a moderate connectivity transition state.

CONCLUSIONS: DFC related to the SN revealed distinct brain network patterns across diagnostic groups, with comorbid PTSD+mTBI showing age- and anxiety-related effects. Older comorbid subjects had heightened hypervigilance and reduced network segregation. PTSD and anxiety may synergistically worsen network instability, while mTBI reflects more rigid, disconnected states, highlighting DFC as a sensitive marker of neuropsychiatric comorbidity.

PMID:40840859 | DOI:10.1016/j.bpsc.2025.08.002

Brain Res. 2025 Aug 19;1865:149896. doi: 10.1016/j.brainres.2025.149896. Online ahead of print.

ABSTRACT

BACKGROUND: Obesity significantly increases not only the incidence and mortality rates of cardiovascular diseases, diabetes, and other metabolic disorders, but also elevates the risk of cognitive impairment-related conditions such as Alzheimer's disease by 3 to 5 times. Based on some brain regions related to reward drive, this study combined brain entropy (BEN) and resting state functional connectivity (RSFC) to explore the neural basis of obesity-induced memory impairment.

METHODS: Based on resting-state functional magnetic resonance imaging (rs-fMRI) data from 23 obese individuals and 36 healthy controls, the BEN values of some brain regions related to reward drive (Nucleus accumbens, Pallidum, Caudate, and Anterior cingulate cortex) were calculated. Mediation analysis was performed to examine whether body mass index (BMI) mediates the relationship between BEN and memory quotient (MQ). Additionally, whole-brain functional connectivity analysis was conducted based on regions showing significant BEN differences.

RESULTS: The BEN in left pallidum (lPAL) was significantly reduced in the obese group compared to controls (p = 0.005). Moreover, BMI mediated the relationship between lPAL entropy and MQ (Indirect effect: 0.2227, 95 % CI [0.0866, 0.3658]). Functional connectivity analysis revealed decreased connectivity between the lPAL and the right superior parietal gyrus, cerebellar Crus II, and cerebellar VIIB in the obese group, along with increased connectivity between the lPAL and the right pallidum.

CONCLUSION: BMI fully mediates the reduced brain complexity of the left pallidum in obese individuals, which is associated with memory impairment and is accompanied by changes in specific functional connectivity patterns. These findings provide new insights into the neural substrates of obesity-related cognitive decline.

PMID:40840855 | DOI:10.1016/j.brainres.2025.149896

Transl Psychiatry. 2025 Aug 21;15(1):303. doi: 10.1038/s41398-025-03500-6.

ABSTRACT

Depression is associated with reduced functional connectivity within the brain's salience network and its strengthened interactions with the default mode network (DMN). Modification of this clinical pattern is challenging. Leveraging the direct neural pathways from olfactory processing regions to the salience network, we explored the effects of electrical stimulation of the olfactory mucosa on brain connectivity. In a randomized, blinded within-subject design, 45 healthy individuals received olfactory or trigeminal nerve stimulation followed by resting-state fMRI. Olfactory stimulation resulted in a significant increase in functional connectivity between the salience network and the piriform cortex - a primary olfactory structure. Importantly, this stimulation increased functional connectivity within the salience network and weakened connectivity between the salience network and the DMN. These findings suggest that olfactory stimulation may modulate connectivity patterns implicated in depression, offering a novel potential minimal invasive therapeutic strategy. However, as these results were obtained from a healthy cohort, further studies are required to evaluate the efficacy in individuals with depression.

PMID:40841359 | PMC:PMC12370952 | DOI:10.1038/s41398-025-03500-6

Comput Biol Med. 2025 Aug 20;196(Pt C):110946. doi: 10.1016/j.compbiomed.2025.110946. Online ahead of print.

ABSTRACT

Understanding the neurobiology of opioid use disorder (OUD) using resting-state functional magnetic resonance imaging (rs-fMRI) may help inform treatment strategies to improve patient outcomes. Recent literature suggests time-frequency characteristics of rs-fMRI blood oxygenation level-dependent (BOLD) signals may offer complementary information to traditional analysis techniques. However, existing studies of OUD analyze BOLD signals using measures computed across all time points. This study, for the first time in the literature, employs data-driven machine learning (ML) for time-frequency analysis of local neural activity within key functional networks to differentiate OUD subjects from healthy controls (HC). We obtain time-frequency features based on rs-fMRI BOLD signals from the default mode network (DMN), salience network (SN), and executive control network (ECN) for 31 OUD and 45 HC subjects. Then, we perform 5-fold cross-validation classification (OUD vs. HC) experiments to study the discriminative power of functional network features while taking into consideration significant demographic features. ML-based time-frequency analysis of DMN, SN, and ECN significantly (p < 0.05) outperforms chance baselines for discriminative power with mean F1 scores of 0.6675, 0.7090, and 0.6810, respectively, and mean AUCs of 0.7302, 0.7603, and 0.7103, respectively. Follow-up Boruta ML analysis of selected time-frequency (wavelet) features reveals significant (p < 0.05) detail coefficients for all three functional networks, underscoring the need for ML and time-frequency analysis of rs-fMRI BOLD signals in the study of OUD.

PMID:40839932 | DOI:10.1016/j.compbiomed.2025.110946

Hum Brain Mapp. 2025 Aug 15;46(12):e70326. doi: 10.1002/hbm.70326.

ABSTRACT

Major depressive disorder (MDD) is a globally prevalent psychiatric disorder that significantly impairs quality of life and increases suicide risk. Accurate identification of MDD is critical for clinically assisted diagnosis. Although substantial progress has been made in MDD identification, extracting region of interest (ROI) features from functional brain networks remains underexplored. Furthermore, most studies rely on small-scale resting-state functional magnetic resonance imaging (rs-fMRI) datasets, which limits the generalizability of their findings to large-scale brain networks. To address these issues, we propose a novel graph embedding-based feature selection classification framework (GEF-FSC) to identify MDD through multi-site rs-fMRI data. The framework employs the node2vec algorithm to learn local and global functional connectivity (FC) features of ROIs via flexible random walks, capturing structural information in functional brain networks. Random Forest is then applied for feature selection on the learned embedding features, followed by classification using an ensemble classifier. This approach captures complex, higher-order structural information between ROIs and retains important features, enhancing classification accuracy by minimizing redundancy in high-dimensional FC features. Evaluated on the REST-meta-MDD dataset, our framework achieved 81.65% accuracy under the Dosenbach template and 75.30% under the AAL atlas. Comparative experiments with eight benchmark methods and six state-of-the-art classifiers demonstrated superior accuracy, sensitivity, specificity, and F1-score. Interpretability analysis highlighted key brain regions and networks consistent with previous findings. The GEF-FSC framework effectively classifies MDD and identifies key brain regions and networks associated with the disorder, emphasizing the importance of higher-order structural information in improving diagnostic accuracy.

PMID:40838619 | PMC:PMC12368899 | DOI:10.1002/hbm.70326

J Psychopharmacol. 2025 Aug 21:2698811251360745. doi: 10.1177/02698811251360745. Online ahead of print.

ABSTRACT

BACKGROUND: Cannabis constituents, including Δ9-tetrahydrocannabinol (THC) and cannabidiol (CBD), show distinct pharmacological profiles with therapeutic relevance for neurological and psychiatric conditions. THC exerts euphoric effects primarily via CB1 receptor activation, while CBD displays non-euphoric properties affecting various pathways.

AIMS: This study evaluated the effects of THC, CBD, and their combination on brain functional connectivity (FC) and cerebral blood flow (CBF) using multimodal neuroimaging.

METHODS: Adult male Sprague Dawley rats received intraperitoneal doses of 10 mg/kg THC, 150 mg/kg CBD, 10.8:10 mg/kg THC:CBD, or vehicle. Resting-state blood oxygenation level dependent magnetic resonance imaging and arterial spin labelling assessed FC and CBF, approximately 2 h after drug administration. Graph-theory metrics and seed-based analyses identified connectivity and perfusion alterations, while plasma analyses determined cannabinoid concentrations.

RESULTS: THC increased whole-brain FC and clustering coefficient, with elevated CBF in cortical and subcortical regions. CBD decreased FC metrics without affecting CBF, while THC:CBD induced moderate increases in both. Seed-based analysis revealed THC-driven increases in cortical-hippocampal and cortical-striatal connectivity, attenuated in the THC:CBD group. A multivariate combined analysis of FC and CBF revealed a divergent pattern of changes induced by each drug.

CONCLUSIONS: In conclusion, we show that THC and CBD induce distinct neurophysiological profiles in rats, with THC increasing both connectivity and perfusion, moderated by CBD when combined. These findings corroborate existing knowledge about the effects of cannabinoids on the brain, while also supporting the potential of preclinical functional neuroimaging to delineate cannabinoid-induced endophenotypes, offering insights for therapeutic development.

PMID:40838351 | DOI:10.1177/02698811251360745

Hum Brain Mapp. 2025 Aug 15;46(12):e70304. doi: 10.1002/hbm.70304.

ABSTRACT

Numerous studies have emphasized the time-varying modular architecture of functional brain networks and its relevance to cognitive functions in healthy participants. However, how modular dynamics of resting-state functional networks change in schizophrenia and how these alterations relate to neurotransmitter and transcriptomic signatures have not been well elucidated. We harmonized resting-state fMRI data from a multi-site sample including 223 patients and 279 healthy controls and applied the multilayer network method to estimate the regional module switching rate (flexibility) of functional brain connectomes. We examined aberrant flexibility in patients relative to controls and explored its relations to neurotransmitter systems and postmortem gene expression. Compared with controls, patients with schizophrenia had significantly higher flexibility in the somatomotor and right visual regions, and lower flexibility in the left parahippocampal gyrus, right supramarginal gyrus, right frontal-operculum-insula, bilateral precuneus, posterior cingulate cortex, and bilateral inferior parietal gyrus. These alterations were associated with multiple neurotransmitter systems and weighted gene transcriptomic profiles. The most relevant genes were preferentially enriched for biological processes of transmembrane transport and brain development, specific cell types, and previously identified schizophrenia-related genes. This study reveals aberrant modular dynamics in schizophrenia and its relations to neurotransmitter systems and schizophrenia-related transcriptomic profiles, providing insights into the understanding of the pathophysiology underlying schizophrenia.

PMID:40838333 | PMC:PMC12368516 | DOI:10.1002/hbm.70304