Most recent paper

Med Image Anal. 2026 Oct 7;115:104350. doi: 10.1016/j.media.2026.104350. Online ahead of print.

ABSTRACT

Neuroimaging-based classification of depression-related symptom and diagnostic groups remains challenging because of clinical and neurobiological heterogeneity. Structural MRI and resting-state functional MRI provide complementary information, but most multimodal approaches rely on global fusion and do not incorporate region-level (ROI) information into cross-modal alignment. We propose a two-stage multimodal framework for depression classification based on ROI-informed cross-modal supervised contrastive learning (RCSCL), which does not require explicit ROI-to-ROI correspondence. Utilizing the Schaefer 200 atlas, we extracted ROI tokens using a MedicalNet-pretrained 3D ResNet-18 for structural MRI and a graph attention network for resting-state fMRI. The tokens were aggregated with attention pooling and the resulting modality-level representations were aligned in a shared embedding space using class-conditional supervised contrastive learning. Global features were then fused using self-attention, and the classification model was optimized using focal loss to address class imbalance. The framework was evaluated on an in-house single-center dataset of 500 subjects and the publicly available SRPBS multi-site MDD dataset. In the primary attention-based implementation, RCSCL significantly increased AUC/AUPRC from 0.805/0.326 to 0.855/0.410 in the in-house dataset and from 0.736/0.704 to 0.766/0.744 on SRPBS, numerically exceeding those of all evaluated fusion baselines without RCSCL; extending RCSCL to these backbones produced further but backbone-dependent gains. Hierarchical visualization analyses identified default mode, limbic, and prefrontal regions as recurrently salient, consistent with prior depression neuroimaging findings, although these attention-derived importance values are model-internal indices. These results suggest that ROI-informed multimodal contrastive learning improves class-discriminative representation learning for multimodal depression classification.

PMID:42868928 | DOI:10.1016/j.media.2026.104350

Psychiatry Res Neuroimaging. 2026 Oct 10;364:112350. doi: 10.1016/j.pscychresns.2026.112350. Online ahead of print.

ABSTRACT

OBJECTIVE: Sex differences may help explain the heterogeneity of internet gaming disorder (IGD), but their neural basis remains unclear. This study examined sex-related alterations in functional connectivity (FC) in IGD.

METHODS: 373 participants (85 males with IGD, 62 females with IGD, 141 male recreational game users [RGUs], and 85 female RGUs) underwent resting-state fMRI. The study used a 2 (group: IGD/RGU) × 2 (sex: male/female) between-subjects design. ReHo and FC were analyzed under this design with age as a covariate. ReHo interaction clusters served as seed ROIs for whole-brain FC analyses, followed by simple effect analyses for significant FC interactions.

RESULTS: ReHo showed significant sex-by-group interactions in the bilateral postcentral gyrus, right middle cingulate cortex, and right fusiform gyrus. Seed-based FC showed significant interactions involving these regions. Simple effects indicated that, relative to same-sex RGUs, males with IGD predominantly showed lower FC, whereas females with IGD more often showed higher FC in connections involving perceptual processing, sensorimotor integration, behavioral regulation, self-related processing, and executive control.

CONCLUSIONS: IGD-related brain functional abnormalities may exhibit sex differences across multiple inter-regional FC patterns, providing new evidence for the sex-related heterogeneity in IGD. Given the cross-sectional design, these findings reflect associations rather than causal effects.

PMID:42868830 | DOI:10.1016/j.pscychresns.2026.112350

Anesth Analg. 2026 Oct 12. doi: 10.1213/ANE.0000000000008327. Online ahead of print.

ABSTRACT

BACKGROUND: Fibromyalgia is a chronic pain disorder frequently co-occurring with depression, anxiety, post-traumatic stress disorder (PTSD), and insomnia, particularly in veterans, and safer nonpharmacological therapies are needed. Cranial electrotherapy stimulation, also referred to as cranial electrotherapy stimulation (CES), may help with anxiety and sleep, but its effect on pain is unclear. This feasibility study evaluated magnetic resonance imaging (MRI) to identify neural correlates of pain and treatment response for CES effects on fibromyalgia pain in veterans.

METHODS: In a randomized, double-blind, sham-controlled trial at a single US government hospital, veterans aged 20 to 60 with fibromyalgia were assigned to active or sham in-home CES, stratified by sex. Participants self-administered CES for 60 minutes daily over 12 weeks. Resting-state functional magnetic resonance imaging (fMRI) was acquired at baseline, 6 weeks, and 12 weeks with clinical outcomes, including the Defense and Veterans Pain Rating Scale (DVPRS), National Institutes of Health (NIH) Patient-Reported Outcomes Measurement Information System (PROMIS) measures, and tests of physical function.

RESULTS: Forty-eight participants were randomized (25 active, 23 sham) and 47 were analyzed. Both groups showed significant improvements in DVPRS pain (-1.88 ± 2.2) and pain interference with sleep (-2.25 ± 2.8), mood (-1.88 ± 2.6), activity (-1.83 ± 2.4) and stress (-1.77 ± 3.0) at 12 weeks versus baseline (P < .01), with significant between-group differences only in sleep (P = .04). Neuroimaging revealed significant increases in connectivity between the cerebellar 7b seed and the left occipital fusiform gyrus (0.079 ± 0.1) left lingual gyrus (0.063 ± 0.08), and right lingual gyrus (0.058 ± 0.08) for the CES group and decreases in the sham group (-0.088 ± 0.1; -0.121 ± 0.08; and -0.135 ± 0.1), respectively. No adverse events occurred.

CONCLUSIONS: Both groups improved similarly in pain and function, though new analgesic usage was more prevalent in the sham group. Neuroimaging revealed differential changes in cerebellar-sensory-visual networks and baseline patterns distinguishing responders from nonresponders, supporting feasibility and design of larger, adequately powered CES trials in fibromyalgia.

PMID:42868517 | DOI:10.1213/ANE.0000000000008327

Front Psychiatry. 2026 Sep 28;17:1944503. doi: 10.3389/fpsyt.2026.1944503. eCollection 2026.

ABSTRACT

BACKGROUND: Traditional auricular therapy for chronic insomnia (Bu Mei in TCM) is used in clinical practice but is limited by insufficient standardization and poorly understood mechanisms. Transcutaneous auricular vagus nerve stimulation (taVNS) provides a quantifiable and reproducible neuromodulatory approach by stimulating vagally innervated regions of the auricular concha, which overlap with auricular acupoints used in TCM. However, evidence in insomnia remains predominantly based on clinical outcomes, whereas longitudinal neuroimaging evidence is limited. Resting-state functional magnetic resonance imaging (rs-fMRI) may complement clinical assessments by characterizing spontaneous brain activity and treatment-related functional changes, thereby providing objective neuroimaging evidence relevant to the potential mechanisms of taVNS.

METHODS: This protocol will recruit 60 patients with chronic insomnia disorder (CID) and 30 age- and sex-matched healthy controls (HCs). Patients with CID will be randomized 1:1 to receive 4 weeks of active or sham-taVNS. Resting-state functional magnetic resonance imaging (rs-fMRI) will be performed at baseline in all participants and repeated at Week 4 in patients with CID, within 24 hours after the final intervention session. Fractional amplitude of low-frequency fluctuation (fALFF) will serve as a prespecified exploratory neuroimaging measure, with amplitude of low-frequency fluctuation (ALFF) included as a complementary exploratory measure. The neuroimaging analyses will examine baseline differences between patients with CID and HCs and between-group differences in baseline-to-Week-4 changes between the active and sham-taVNS groups to characterize brain functional alterations associated with CID and changes associated with active taVNS relative to sham-taVNS. The primary clinical outcome will be the change in Pittsburgh Sleep Quality Index (PSQI) score from baseline to Week 4. Secondary and follow-up assessments will evaluate insomnia severity, daytime sleepiness, emotional status, cognitive function, pre-sleep arousal, and quality of life. Exploratory correlations between changes in neuroimaging measures and clinical outcomes will also be examined.

DISCUSSION: This study will integrate clinical assessments with rs-fMRI to evaluate the clinical effects of taVNS in patients with CID and explore associated changes in spontaneous brain activity. The integration of clinical outcomes and exploratory neuroimaging measures may provide complementary evidence on treatment effects and associated brain functional changes, thereby informing future investigation of the potential neural mechanisms of taVNS in CID.

CLINICAL TRIAL REGISTRATION: https://www.chictr.org.cn/hvshowprojectEN.html?id=277442&v=1.0, identifier ChiCTR2500103157.

PMID:42865784 | PMC:PMC13655151 | DOI:10.3389/fpsyt.2026.1944503

Mov Disord. 2026 Oct 12. doi: 10.1002/mds.70552. Online ahead of print.

ABSTRACT

BACKGROUND: Deep brain stimulation (DBS) is an established treatment for craniocervical dystonia (CCD), yet postoperative benefit varies widely and cannot be estimated before surgery. Resting-state functional MRI (rs-fMRI) has shown prognostic value across DBS indications. The objective of this study was to develop and externally validate preoperative rs-fMRI models estimating DBS response in CCD.

METHODS: Forty-two patients from two independent centers underwent preoperative 3 T rs-fMRI (discovery, n = 26; validation, n = 16). Using 12 literature-informed, symptom-relevant regions of interest, we quantified the amplitude of low-frequency fluctuation as a measure of intrinsic regional neural activity. We applied regularized modeling to classify super-responders, distinguish responders from non-responders, and estimate improvement in Burke-Fahn-Marsden Dystonia Rating Scale motor subscore.

RESULTS: When applied to the validation cohort, the imaging models achieved a three-category classification accuracy of 75.0% (50.5%-89.8%) using a sequential two-step procedure (Pperm = 0.002). Predicted continuous improvement was also correlated with empirical improvement (Rext = 0.65, Q2 = 0.34, Pperm = 0.004). Within the discovery cohort, nested leave-one-out-cross-validation, in which feature selection and decision-threshold derivation were repeated independently within each training fold, yielded areas under the curve (AUCs) of 0.73 (0.50-0.96) and 0.77 (0.51-1.00). Across models, baseline primary motor cortex activity was consistently associated with diminished postoperative benefit.

CONCLUSIONS: Preoperative rs-fMRI captures clinically relevant variation in baseline brain state in CCD, with predictive performance supported in an external validation. These findings provide preliminary support for rs-fMRI as a potential source of information to aid patient counseling and clinical decision-making, pending prospective validation. © 2026 International Parkinson and Movement Disorder Society.

PMID:42864990 | DOI:10.1002/mds.70552

IEEE Trans Med Imaging. 2026 Oct 12;PP. doi: 10.1109/TMI.2026.3742915. Online ahead of print.

ABSTRACT

Conventional EPI-based fMRI relies on the BOLD contrast to map activity changes, but is limited by high acoustic noise, susceptibility artifacts, and distortion near air-tissue interfaces. Ultra-short and zero-echo-time (ZTE) sequences have gained considerable attention in animal fMRI studies due to their robustness against susceptibility effects, low acoustic noise, and ability to capture signal sources distinct from those of conventional BOLD imaging. In this study, we present a feasibility-level proof-of-concept technical demonstration of ZTE-based fMRI in humans at 3T, while maintaining many of the advantages observed in preclinical studies. Phantom measurements demonstrate that the ZTE protocol is sensitive to oxygenation-related T1 changes at 3T, supporting its feasibility for detecting related functional signals. In five healthy volunteers, task-based visual stimulation and finger tapping experiments elicited activation in primary visual and motor cortices, respectively, using the ZTE sequence, demonstrating proof-of-concept feasibility in this small cohort. Resting-state components with spatial patterns consistent with canonical visual and sensorimotor networks were identified using independent component analysis in the same cohort. Compared to BOLD-EPI, ZTE showed significantly reduced acoustic noise and was largely insensitive to susceptibility-induced signal loss in this feasibility study. While this work represents our initial experience using ZTE for human fMRI at 3T, the presented results will motivate future studies to refine and extend this approach to explore complementary functional contrast mechanisms.

PMID:42862611 | DOI:10.1109/TMI.2026.3742915

Dialogues Clin Neurosci. 2027 Dec;29(1):1-14. doi: 10.1080/19585969.2026.2742156. Epub 2026 Oct 12.

ABSTRACT

BACKGROUND: Although treatment approaches have been explored in internet gaming disorder (IGD), however, little is known about the specific neural activity that is associated with treatment outcomes. This study aims to establish a treatment-sensitive network that reliably predicts treatment outcomes across different interventions for IGD.

METHODS: Resting-state fMRI data were collected from 90 individuals with IGD who received one of three different intervention protocols (mindfulness meditation, approach bias modification, or retrieval-aversive conditioning training). Connectome-based predictive modelling (CPM) was used to build a treatment-sensitive network to predict individual symptom reduction, then validated them on the external dataset.

RESULTS: The negative (r = 0.401, p <.001) and combined network (r = 0.480, p <.001) can significantly predict a reduction of addiction severity (baseline minus post-test). Specifically, the negative network was primarily centred on the motor, subcortical, and salience networks. The positive network primarily involved cerebellar-related connections. Incorporating these into a combined model significantly improved predictive accuracy, suggesting that cerebellar-related connectivity may contribute to the prediction of intervention-related clinical improvement. Finally, the treatment-sensitive network was validated in a separate data.

CONCLUSIONS: This study established a treatment-sensitive network based on the whole-brain connectome. These findings provide valuable empirical evidence for identifying biomarkers for IGD treatment and advance personalised therapeutic strategies.

CLINICAL TRIALS REGISTRATION: The trial protocol was pre-registered with the Chinese Clinical Trial Registry (www.chictr.org.cn; ChiCTR2300077831(MM), ChiCTR2300075869(ApBm), ChiCTR2500095444(R-AC), ChiCTR2400090834(SAE)).

PMID:42862599 | DOI:10.1080/19585969.2026.2742156

Quant Imaging Med Surg. 2026 Oct 1;16(10):766. doi: 10.21037/qims-2026-0906. Epub 2026 Sep 3.

ABSTRACT

BACKGROUND: Mild pain can influence brain function, yet its central neurobiological mechanisms in chronic low back pain (CLBP) remain insufficiently defined. This study aimed to quantitatively evaluate localized and long-range functional connectivity (FC) alterations in patients with mild to moderate CLBP using static/dynamic regional homogeneity (sReHo/dReHo) and seed-based FC, and to assess their statistical relationships with clinical pain and psychological metrics.

METHODS: A total of 55 mild to moderate CLBP patients and 39 healthy controls (HCs) underwent 3.0 T resting-state functional magnetic resonance imaging (rs-fMRI) and clinical assessments, including Visual Analogue Scale (VAS), Short-Form of the McGill Pain Questionnaire (SF-MPQ), Self-Rating Anxiety Scale (SAS), Self-Rating Depression Scale (SDS), and Pittsburgh Sleep Quality Index (PSQI). sReHo and dReHo (sliding window =50 TRs, step size =5 TRs) maps were computed to identify abnormal localized brain regions. Brain regions showing significant group differences were selected as seeds (6-mm spherical regions of interest, ROIs) for whole-brain FC analysis. Statistical comparisons were corrected using Gaussian random field (GRF) theory (voxel-level P<0.005, cluster-level P<0.05). Partial correlation analyses were used to control for age, sex, education, and head motion (Bonferroni-corrected threshold P<0.007).

RESULTS: Compared with HCs, patients exhibited significantly reduced sReHo in the third cerebellar vermis (peak t=-4.253, GRF P<0.05), alongside reduced dReHo in the right parahippocampal gyrus (t=-4.796, GRF P<0.05) and left inferior temporal gyrus (t=-3.935, GRF P<0.05). Seed-based FC analysis revealed decreased connectivity between the third cerebellar vermis and right hippocampus (t=-4.678, GRF P<0.05), and increased connectivity between the left inferior temporal gyrus and left pallidus (t=4.441, GRF P<0.05). Patients showed significantly higher SAS anxiety scores (25.15±5.01 vs. 23.08±2.18, P=0.008). Correlation analyses demonstrated that sReHo in the third cerebellar vermis was negatively correlated with SAS scores (r=-0.384, P=0.005), whereas FC between the left inferior temporal gyrus and left pallidus was positively correlated with SF-MPQ sensory (r=0.381, P=0.006), affective (r=0.386, P=0.005), and total scores (r=0.445, P=0.001).

CONCLUSIONS: Patients with mild to moderate CLBP exhibit distinct localized and long-range FC alterations within cerebello-hippocampal and cortico-basal ganglia pathways. These neurofunctional changes may contribute to the integration of subjective pain dimensions and comorbid anxiety. Combining static/dynamic ReHo with seed-based FC provides complementary neurobiological insight into how low-intensity chronic pain may affect the central nervous system and highlights the importance of early clinical recognition.

PMID:42861121 | PMC:PMC13654137 | DOI:10.21037/qims-2026-0906

Quant Imaging Med Surg. 2026 Oct 1;16(10):814. doi: 10.21037/qims-2026-1069. Epub 2026 Sep 15.

ABSTRACT

BACKGROUND: Whether hearing loss fundamentally alters tinnitus-related brain reorganization remains unclear. To address this issue, this study compared features from multimodal resting-state functional magnetic resonance imaging (rs-fMRI) between tinnitus patients with hearing loss (TIN + HL) and those without hearing loss (TIN - HL).

METHODS: A total of 38 patients with tinnitus (23 TIN + HL and 15 TIN - HL) and 28 healthy controls (HCs) underwent rs-fMRI. Amplitude of low-frequency fluctuations (ALFF), regional homogeneity (ReHo), and bilateral auditory cortex (AC)-based functional connectivity (FC) were compared among groups via one-way analysis of variance (ANOVA) and post hoc t tests. Correlations with clinical characteristics were examined via partial correlation.

RESULTS: Compared with the HC group, both tinnitus groups exhibited decreased ALFF in the bilateral anterior cingulate cortex (ACC), along with increased ALFF and decreased ReHo in the left middle occipital gyrus (MOG) [cluster-level family-wise error (FWE)-corrected P<0.05]. Additionally, the TIN + HL group showed decreased ALFF in the bilateral caudate nucleus (CN) and decreased ReHo in the right parahippocampal gyrus (PHG) and middle frontal gyrus (MFG), while the TIN - HL group exhibited decreased ReHo in the right rectus gyrus. Compared with the HC group, both tinnitus groups showed reduced FC between the right AC and the left superior temporal gyrus (STG), with more pronounced reduction in the TIN - HL group (P<0.05, Bonferroni correction). The ALFF in left MOG negatively correlated with the depression scores in the tinnitus groups (r=-0.513; P=0.001).

CONCLUSIONS: Hearing loss may differentially shape tinnitus-related brain reorganization. Although both subtypes shared common alterations in limbic and visual association regions, TIN + HL appears to be characterized by disruptions in sensory gating and cognitive control circuits, whereas TIN - HL may involve emotion-regulation network alterations and more pronounced auditory network disconnection. These findings provide a neuroimaging basis for subtype-specific treatment.

PMID:42861109 | PMC:PMC13654167 | DOI:10.21037/qims-2026-1069

Neuroimage Rep. 2026 Sep 30;6(4):100415. doi: 10.1016/j.ynirp.2026.100415. eCollection 2026 Dec.

ABSTRACT

Mapping functional brain organization between species is a prerequisite for translating connectivity biomarkers from animal models to patients, but connectivity-based correspondence has been demonstrated almost exclusively in primates. We tested whether canine-human network homology can be recovered from functional connectivity alone, using only five network-level anatomical priors, and benchmarked a spectral manifold alignment framework (Laplacian-eigenmap embedding with Procrustes alignment on network centroids) against a simple connectivity-signature matching baseline. Data comprised resting-state fMRI from 15 canines with endovascular permanent middle cerebral artery occlusion (pre-occlusion baseline and hyperacute post-occlusion sessions) and two public OpenNeuro datasets: 29 chronic-phase post-stroke aphasia patients scanned before and after rehabilitation (ds003999) and 10 healthy controls (ds000224). Held-out canine-human network homologies were recovered well above chance (top-5 network correspondence up to 54.5% vs. permutation null ≈ 18%, p = 0.002 ), and correspondence followed the evolutionary cortical hierarchy: visual and somatomotor homologies were recovered several-fold above chance, default-mode only weakly-extending a gradient previously described in primates to a carnivore. Notably, simple signature matching (54.5%) outperformed manifold alignment (35.5%), indicating that the homology signal is strong enough to be captured without embedding machinery at translational cohort sizes. Within species, canine network metrics showed no treatment-arm differences (all p > 0.3 ), and chronic-phase patients did not differ from controls in modularity ( p = 0.93 ); a sensitivity analysis bounds these nulls. All results are fully reproducible from shared code and public data.

PMID:42861075 | PMC:PMC13654029 | DOI:10.1016/j.ynirp.2026.100415

Neuroimage Clin. 2026 Oct 8;52:104063. doi: 10.1016/j.nicl.2026.104063. Online ahead of print.

ABSTRACT

BACKGROUND: Post-stroke fatigue is common and disabling, yet rarely distinguished from co-occurring depression, limiting understanding of its neural substrates. We examined resting-state fMRI correlates of post-stroke fatigue while accounting for depressive symptoms.

METHODS: We studied 103 stroke patients and 29 age-matched controls from a longitudinal neuroimaging cohort. Fatigue was assessed via the SF-36 Vitality subscale and depression via the Geriatric Depression Scale - Short (GDS-15). Linear mixed-effects models evaluated associations between vitality and resting-state functional connectivity, adjusting for depression and covariates. Residualized vitality models served as sensitivity analyses to isolate fatigue variance independent of depressive symptoms, with FDR correction applied within feature families.

RESULTS: In depression-adjusted models, higher vitality (lower fatigue) was associated with increased frontoparietal-limbic connectivity (β = 0.0013, FDR q = 0.048), though this did not survive depression residualization. In a residualized sensitivity analysis, higher vitality (lower fatigue) was associated with reduced thalamus-insula and thalamus-ventral attention network connectivity (thalamus-insula: β = -0.0071, FDR q = 0.011; thalamus-VAN: β = -0.0043, FDR q = 0.012); these associations were attenuated and did not survive correction in the primary depression-adjusted model (q ≈ 0.075).

CONCLUSIONS: Fatigue was associated with both frontoparietal-limbic and thalamocortical connectivity, although the thalamic associations were attenuated after adjustment for depressive symptoms in the primary analysis and were most evident in residualized sensitivity analyses. These findings suggest that fatigue and depressive symptoms may involve both shared and potentially distinct neural correlates. Point estimates were smaller in stroke participants than controls, providing no evidence that the observed associations were stroke-specific. Future work should determine whether these relationships are unique to stroke or reflect broader neural correlates of fatigue across clinical populations.

PMID:42858620 | DOI:10.1016/j.nicl.2026.104063

Mov Disord. 2026 Oct 10. doi: 10.1002/mds.70564. Online ahead of print.

ABSTRACT

BACKGROUND: The cerebellum is increasingly recognized as a key contributor to cognitive reserve and network organization in Parkinson's disease (PD). However, the relationship of cerebello-cortical and cerebello-basal ganglia connectivity with cognitive status and nigrostriatal integrity remains incompletely understood.

METHODS: Resting-state functional magnetic resonance imaging (fMRI) data from 172 participants with PD in the Parkinson's Precision Medicine Initiative were analyzed. Cerebello-basal ganglia and cerebello-cortical connectivity were investigated using region of interest (ROI)-to-ROI and seed-to-voxel approaches. Cerebellar connectivity was examined in relation to cognitive status, age, disease duration, motor severity, dopaminergic medication exposure, and contralateral putaminal striatal binding ratio (SBR).

RESULTS: PD participants demonstrated robust positive cerebello-basal ganglia connectivity, particularly between lobule VI/vermis VI and the pallidum. Greater age was associated with widespread reductions in cerebello-basal ganglia coupling. Longer disease duration was associated with stronger connectivity within cerebellar and basal ganglia circuits, and among cognitively unimpaired participants, stronger lobule V connectivity with temporal regions. After adjustment for contralateral putaminal SBR, cognitively unimpaired participants generally demonstrated greater connectivity between left cerebellar regions and frontal networks, whereas cognitively impaired participants showed greater connectivity from posterior cerebellar and vermal regions, including regions implicated in cognitive processing, to temporal, parahippocampal, occipital, parietal, and somatosensory regions.

CONCLUSIONS: Cerebellar functional connectivity in PD shows region-specific differences across cognitive status rather than uniform network loss with cognitive impairment. These findings support further investigation of cerebellar functional connectivity as a candidate biomarker of cognitive resilience and network adaptation in PD. © 2026 The Author(s). Movement Disorders published by Wiley Periodicals LLC on behalf of International Parkinson and Movement Disorder Society.

PMID:42857890 | DOI:10.1002/mds.70564

Cogn Neurodyn. 2026 Dec;20(1):198. doi: 10.1007/s11571-026-10568-3. Epub 2026 Oct 8.

ABSTRACT

Repetitive transcranial magnetic stimulation (rTMS) is a proven tool to modulate brain functional dynamics and treat neurological and psychiatric diseases. However, it remains unclear how local rTMS affects whole-brain functional dynamics. Here, we investigated the stimulation effects of rTMS using two data-driven approaches. By constructing multi-layer networks from resting-state fMRI data acquired before and after 1 Hz stimulation over the left frontal (FRO-TMS) and occipital cortex (OCC-TMS), we found significantly decreased dynamic interaction after FRO-TMS. We further characterized whole-brain functional hierarchy by computing nonreversibility and observed a significantly flattened hierarchy after FRO-TMS, whereas OCC-TMS did not induce significant changes. We then systematically investigated the association between these distinct modulatory patterns and structural network organization by fitting a large-scale brain computational model. Using noise- and synchronization-based perturbation protocols, we found that the simulated perturbation distance correlated significantly with the local dominance and controllability of the stimulated node in the structural network, providing a potential mechanistic account for the empirical observations. Specifically, driving hub nodes induced relatively small perturbation distances, whereas driving weakly connected regions yielded larger distances. This study provides a heuristic strategy for modeling and optimizing stimulation protocols in noninvasive brain stimulation.

SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at https://doi.org/10.1007/s11571-026-10568-3.

PMID:42857245 | PMC:PMC13650214 | DOI:10.1007/s11571-026-10568-3

Cogn Neurodyn. 2026 Dec;20(1):199. doi: 10.1007/s11571-026-10563-8. Epub 2026 Oct 8.

ABSTRACT

Autism spectrum disorder (ASD) has resisted reproducible case-control characterization in resting-state functional connectivity, leaving open whether reported group differences reflect genuine biology or methodological variability. We introduce a geometric framework that represents functional connectomes on the symmetric positive-definite Riemannian manifold and quantifies individual deviation as tangent-space dispersion, complemented by normative modeling of graph topology. Using the multi-site Autism Brain Imaging Data Exchange (ABIDE I) cohort (N=1035 subjects), we evaluate this descriptor under stringent robustness conditions and report three findings. First, individual connectome geometry is a stable, individually unique descriptor. The magnitude of a subject's deviation is highly reliable across split halves (intraclass correlation ICC[Formula: see text]) but is dominated by an acquisition-site scale, whereas its high-dimensional shape identifies subjects within-session with 98.9% accuracy once that scale is removed. Second, the case-control difference in dispersion is null and, by two one-sided equivalence tests against Cohen's d = 0.2, statistically equivalent to zero in all ABIDE I analysis cohorts (full cohort d = -0.007, N = 1035; strict cohort d = +0.038, N = 635); a nominal single-atlas effect (dispersion ratio 1.05, p = 0.046) does not survive multi-atlas reanalysis. Third, acquisition site explains 69% of dispersion variance while diagnosis explains 0.0001%; deviations form no discrete subtypes and do not predict symptom severity (minimum p=0.15 across tested graph-deviation predictors). Geometric dispersion is therefore a reproducible, individually unique connectome descriptor but not a case-control biomarker of ASD, which helps to explain the inconsistency of connectomic findings in autism.

PMID:42857201 | PMC:PMC13650096 | DOI:10.1007/s11571-026-10563-8

medRxiv [Preprint]. 2026 Oct 1:2026.09.29.26364036. doi: 10.64898/2026.09.29.26364036.

ABSTRACT

BACKGROUND: Response to Comprehensive Behavioral Intervention for Tics (CBIT), the first-line treatment for tic disorders, is suboptimal. We conducted Phase 1 of a two-phase, milestone-driven trial to determine if augmenting CBIT with repetitive transcranial magnetic stimulation (TMS) to supplementary motor area (SMA) could safely engage targeted sensorimotor circuitry and improve tic controllability.

METHODS: In a 3-arm, double-blind, sham-controlled randomized trial ( NCT04578912 ), 50 youth (ages 12-21) with moderate-to-severe tics received 10 sessions of either 1Hz rTMS, continuous theta burst stimulation (cTBS), or sham stimulation to the SMA, immediately followed by CBIT. Primary outcomes were safety, feasibility, and neural target engagement (fMRI SMA activation and resting-state functional connectivity [RSFC]). Secondary outcomes included change in tic controllability, tic frequency at rest, and urges.

RESULTS: The CBIT+TMS intervention was safe and feasible (96% completion). Treatment-related adverse events were mild-to-moderate and occurred more frequently in the 1Hz rTMS group (27% of sessions) vs. cTBS (14.7%) and sham (11.3%). While task-based SMA activation remained unchanged, cTBS demonstrated robust neural target engagement, reducing SMA-primary motor cortex (M1) RSFC and putamen-M1 RSFC. Conversely, 1Hz rTMS blunted the cortico-striatal connectivity reductions observed in the sham group. All groups exhibited significant reductions in tic frequency and improved controllability. Premonitory urges increased in the 1Hz rTMS group but decreased in sham and cTBS.

CONCLUSIONS: SMA-directed cTBS is a safe, feasible augmentation that may amplify the neural effects of CBIT on sensorimotor circuitry. Findings identify cTBS as the preferred stimulation regimen for future efficacy testing.

CLINICAL TRIAL REGISTRATION: ClinicalTrials.gov; Identifier: NCT04578912 .

PMID:42856611 | PMC:PMC13648794 | DOI:10.64898/2026.09.29.26364036

bioRxiv [Preprint]. 2026 Sep 30:2026.09.24.754231. doi: 10.64898/2026.09.24.754231.

ABSTRACT

Resting-state connectivity can predict task-evoked fMRI activation, but correspondence with an individual task map may partly reflect a shared population pattern. We evaluated the Variational Resting-state-to-Task Prediction TransformeR (VRPTR), a three-dimensional encoder-decoder combining a compressed Transformer bottleneck, variational latent sampling, and multiscale skip connections. Models were trained in 360 healthy adults from the WU-Minn Human Connectome Project and evaluated in 40 held-out participants from the same dataset for the story-versus-math language contrast. VRPTR achieved mean voxel-map Pearson r=0.642 and Dice AUC=0.519, exceeding compact volumetric BrainSurfCNN-like and SWIFUN-like comparators by Δr=0.0376 and 0.0335, respectively. A template fixed from the 360 training maps yielded residual correlation 0.364; VRPTR improved correlation but not mean absolute or root mean squared error over this template. A matched factorial ablation associated Transformer processing with improved map correlation (Δr=0.0168); the variational main effect did not survive multiple-comparison correction. Component swaps indicated that subject-specific activation patterns depended more on multiscale skip features than on the bottleneck representation. Latent sampling produced a modest association between posterior standard deviation and voxelwise absolute error (mean Spearman ρ=0.091). Raw 95% intervals covered only 4.7% of observed values. Five-fold calibration within the held-out cohort increased coverage to 94.8%, and spatially varying uncertainty modestly improved probabilistic scores over a constant-width model. Uncertainty also increased under input degradation. These findings support the contribution of Transformer processing to language-map prediction and distinguish it from subject-specific spatial conditioning and relative uncertainty estimation. Independent validation of calibrated uncertainty and evaluation in clinical cohorts remain necessary.

PMID:42856173 | PMC:PMC13648463 | DOI:10.64898/2026.09.24.754231

bioRxiv [Preprint]. 2026 Sep 29:2026.09.23.753381. doi: 10.64898/2026.09.23.753381.

ABSTRACT

Neuroimaging is becoming an increasingly prominent tool to understand intimate partner violence (IPV) and partner-inflicted brain injury (BI). Yet, by nature of IPV and partner-inflicted BI, those most affected may be the least likely to be eligible, participate, and produce usable data in neuroimaging studies. A critical question therefore remains of who is represented in neuroimaging research on IPV and related BI. Here, we examined representativeness at three major stages of neuroimaging sample construction using screening records from 352 women and a subsequent cohort of 207 enrolled women with lifetime exposure to physical IPV. At the eligibility stage, three in five women eligible for behavioral research were deemed ineligible for magnetic resonance imaging (MRI) at initial screening. At the participation stage, the 61 women who underwent resting-state functional MRI (fMRI) had more education, were more likely to be employed, and reported lower psychiatric and neurobehavioral symptom burden than enrolled women who did not undergo neuroimaging. Finally, at the quality-control stage following fMRI, motion-based exclusion disproportionately removed women who had sustained the most partner-inflicted BIs. Our findings empirically reveal that eligibility, participation, and quality control each bias the demographic, clinical, abuse, and BI burden represented in an fMRI sample, and therefore have key implications for generalizability. Measuring and reporting representation bias can help identify whose experiences are underrepresented in the evidence, and accordingly reduce the risk that IPV neuroimaging research understates the very harms it seeks to characterize.

PMID:42856107 | PMC:PMC13648376 | DOI:10.64898/2026.09.23.753381

Mol Psychiatry. 2026 Oct 9. doi: 10.1038/s41380-026-03921-9. Online ahead of print.

ABSTRACT

Psychosis exhibits marked neurobiological heterogeneity that is not adequately captured by symptom-based diagnoses. Although prior efforts have identified subgroups, it remains unclear how latent disease variability manifests within the brain's multiscale intrinsic functional organization and whether these patterns can be leveraged to systematically identify biologically driven subtypes of psychosis. Here, we aim to identify Psychosis Imaging Neurosubtypes (PINs)-homogeneous subgroups defined solely from neurobiological features of brain function, independent of clinical or cognitive definitions, and whether these subgroups show meaningful phenotypic differentiation. Resting-state fMRI data from 2103 B-SNIP 1&2 participants were used to compute subject-specific multiscale functional network connectivity (msFNC). We then derived a low-dimensional neurobiological subspace, termed Latent Network Connectivity, which captured system-wide interconnected multiscale information across three components. Projections of probands msFNC onto this subspace revealed three PINs through unsupervised learning that spanned DSM diagnoses (Schizophrenia, Bipolar, Schizoaffective Disorder). Although defined purely from intrinsic neurobiology, PINs exhibited distinct cognitive, clinical, and connectivity profiles. PIN-1, the most cognitively impaired, showed Cerebellar-Subcortical and Visual-Sensorimotor hypoconnectivity, alongside Visual-Subcortical hyperconnectivity. Most cognitively preserved PIN-2 showed Visual-Subcortical, Subcortical-Sensorimotor, and Subcortical-Higher Cognition hypoconnectivity. PIN-3 exhibited intermediate cognition, showing Cerebellar-Subcortical hypoconnectivity alongside Cerebellar-Sensorimotor and Subcortical-Sensorimotor hyperconnectivity. Although covariate-adjusted analyses did not reveal significant clinical differences among PINs, exploratory analyses identified selective differences in YMDR, PANSS, and Age of Onset. Notably, 55% of relatives aligned with the same neurosubtype as their probands, a rate significantly higher than chance and than DSM-based alignment, suggesting potential heritability. Cognitive performance reliably aligns with distinct brain connectivity patterns, which are also evident in relatives, supporting their construct validity. Our PINs partially converged with previously reported BSNIP Biotypes with some correspondence (e.g. PIN1 with Biotype 1) but not with the other Biotypes, which were determined using integrated electrophysiological, cognitive, and oculomotor data, supporting convergent validity, while also indicating that intrinsic rsfMRI derived multiscale connectivity partitions psychosis heterogeneity somewhat differently from multimodal biomarker approaches. These findings underscore the limitations of DSM-based classifications in capturing the biological complexity of psychotic disorders and highlight the potential of imaging-based neurosubtypes to enhance our understanding of the psychosis spectrum.

PMID:42855456 | DOI:10.1038/s41380-026-03921-9

Brain Struct Funct. 2026 Oct 9;231(8):149. doi: 10.1007/s00429-026-03203-8.

ABSTRACT

The rostromedial tegmental nucleus (RMTg) has been shown to inhibit midbrain dopaminergic centers through its GABAergic projections, and to be critical for regulating responses to aversive and rewarding stimuli. This nucleus has primarily been evaluated in rodents and non-human primates, while recent postmortem analysis of human brainstem tissue localized GABAergic cell clusters in the putative RMTg. The objectives of this preliminary investigation in healthy controls (HC) were to (i) define RMTg function during evoked thermal stimulation and the resting-state using 7T functional magnetic resonance imaging (fMRI), and (ii) explore the expression of putative RMTg markers, µ-opioid receptors (OPMR1), nociceptin receptors (NOPR), and FOXP1 as well as tyrosine hydroxylase (TH) in the midbrain tegmentum. Relative to 40 °C stimuli, greater responses to 46 °C (normally noxious) in the putative RMTg as well as raphe nuclei, locus coeruleus, substantia nigra, thalamus, putamen, and anterior cingulate were observed. Resting-state 7T fMRI showed robust RMTg functional connectivity with components of mesocorticolimbic and nigrostriatal circuitry, thalamus, hippocampus, and cerebellar subdivision. In postmortem analyses, neurons immunolabeled for OPMR1, NOPR, and FOXP1 were detected in an area adjacent to the decussation of the superior cerebellar peduncle. These immune-positive neurons in the area of the putative RMTg were adjacent to intense TH immunoreactivity. This investigation provides novel evidence of activation of a distinct area within the human midbrain tegmentum, putatively identified as the RMTg, particularly during nociceptive processing in addition to its resting-state connectivity with the broader central nervous system. Further exploration of the human RMTg is warranted.

PMID:42853412 | DOI:10.1007/s00429-026-03203-8

Br J Radiol. 2026 Oct 9:tqag229. doi: 10.1093/bjr/tqag229. Online ahead of print.

ABSTRACT

OBJECTIVES: To investigate alterations in CBF-ALFF spatial coupling in obstructive sleep apnoea (OSA).

METHODS: Thirty-two patients with OSA and 32 healthy controls（HCs） matched for sex, age, and education background were recruited for this work. All participants received arterial spin labelling (ASL) and resting-state functional MRI (rs-fMRI) examinations. Cerebral blood flow (CBF) and amplitude of low-frequency fluctuation (ALFF) maps were derived from ASL and fMRI data, respectively. Between-group comparisons were performed to identify differences in ALFF, CBF, and CBF-ALFF coupling metrics. Subsequently, correlation analyses were conducted in OSA patients to investigate whether CBF/ALFF ratios derived from brain regions exhibiting significant group-level differences were associated with polysomnographic and clinical parameters.

RESULTS: Patients with OSA exhibited reduced global spatial CBF-ALFF coupling relative to HCs. Regionally, decreased CBF/ALFF ratios were primarily observed in the left parahippocampal gyrus, whereas increased ratios were located mainly in the default mode network (DMN), including the left precuneus, right angular gyrus, and left inferior parietal lobule.

CONCLUSIONS: This study demonstrated disrupted spatial CBF-ALFF coupling in OSA patients. Such coupling alterations reflect broad OSA-associated pathophysiological changes, including but not limited to neurovascular dysfunction, offering new insights into the neuropathological mechanisms of OSA.

ADVANCES IN KNOWLEDGE: Using combined BOLD and ASL imaging, this study provides the first characterization of both global and regional CBF-ALFF coupling alterations in OSA, revealing distinct uncoupling signatures in the DMN and medial temporal lobe that deepen our understanding of OSA-related neurovascular pathophysiology.

PMID:42853094 | DOI:10.1093/bjr/tqag229