Maged Goubran

THE BIGGER PICTURE: We now have imaging and analytical tools to study the brain at multiple scales, from scanning the whole brain to see how different neuronal networks interact to scrutinizing individual cells. However, multimodal and multiscale investigations are challenging to conduct in the same subject and generate copious amounts of complex data. Such experimental designs are invaluable in studying heterogeneous neurological disease with pathology at both cellular and network level. Using traumatic brain injury as a test case, we conducted macroscale MRI imaging as well as 3D whole-brain light-sheet fluorescence microscopy with cellular resolution to study network dysfunction and markers of cognitive deficits post-injury. We identified sensitive neurophysiological markers that distinguish healthy and injured mice, and predicted their memory capabilities. The framework we present is broadly applicable to understanding complex cognitive functions and identifying the physiological correlates of distributed network dysfunction in neurological disorders. HIGHLIGHTS: ● Multiscale framework integration of in vivo MRI and ex vivo whole-brain histology ● Connectomic gradients reveal network dedifferentiation linked to memory deficit ● 3D whole-brain segmentation show neuronal hyperactivity and widespread astrogliosis ● AI-based axon tracing reveals persistent damage despite normal neuroimaging scans SUMMARY: Integrating high-dimensional neuroimaging data across spatial scales remains a fundamental bottleneck in quantitative neuroscience. Here, we apply emerging analytical techniques including functional connectome gradient mapping and 3D segmentation of whole-brain cellular resolution microscopy data in a mouse model of traumatic brain injury. At the macroscale, longitudinal functional MRI revealed a transition from frontal hyperconnectivity to hippocampal hypoconnectivity and unimodal-transmodal gradient contraction indicating network dedifferentiation. At the cellular scale, deep-learning based segmentation of whole-brain light-sheet fluorescence microscopy volumes quantified regional astrogliosis, neuronal hyperactivity, and single-axon loss undetected by standard diffusion MRI. Multiscale multimodal data integration not only identified quantitative biomarkers predictive of memory performance post-injury but also enabled validation of in vivo modalities against more sensitive ex vivo histology. This study demonstrates a framework for probing complex multi-network cognitive functions and neurological disorders of network dysfunction.

Integrating high-dimensional neuroimaging data across spatial scales remains a fundamental bottleneck in quantitative neuroscience. Here, we apply emerging analytical techniques including functional connectome gradient mapping and 3D segmentation of whole-brain cellular resolution microscopy data in a mouse model of traumatic brain injury. At the macroscale, longitudinal functional MRI revealed a transition from frontal hyperconnectivity to hippocampal hypoconnectivity and unimodal-transmodal gradient contraction indicating network dedifferentiation. At the cellular scale, deep-learning based segmentation of whole-brain light-sheet fluorescence microscopy volumes quantified regional astrogliosis, neuronal hyperactivity, and single-axon loss undetected by standard diffusion MRI. Multiscale multimodal data integration not only identified quantitative biomarkers predictive of memory performance post-injury but also enabled validation of in vivo modalities against more sensitive ex vivo histology. This study demonstrates a framework for probing complex multi-network cognitive functions and neurological disorders of network dysfunction.

Background: Preexisting factors are among the strongest predictors of recovery following traumatic brain injury (TBI), with chronic stress closely associated with permanent disability and worse long-term outcomes. While chronic cerebrovascular dysfunction is linked to these poor trajectories, as impaired blood flow regulation drives secondary disease progression, the mechanisms regulating this vascular failure remain incompletely understood. Crucially, how premorbid chronic stress and TBI disrupt the fundamental coordination of the cerebrovascular network post-injury remains unknown. Methods: To elucidate this coordination at rest and in response to increased neuronal activity, we used a model of three repeated moderate closed cortical impacts comorbid with chronic stress induced by social isolation (SI) post-weaning. Our previously developed vascular analysis pipeline (NOVAS3D) was employed to estimate changes in vascular radii across cerebrovasculature proximal to neuronal activation. Arteries and veins were annotated in the imaged volumes to allow for blood flow simulations. Results: Using graph-based network analysis, we demonstrate that TBI, when compounded with chronic stress, critically disrupts the long-range coordination of the capillary network. Specifically, the functional coordination of radius changes between nearby, non-adjacent capillaries was reduced by 40±20% in TBI+SI mice relative to controls. Consequently, simulations estimated that the vascular networks in TBI±SI mice experienced a 68±7% reduction in arterial red blood cell velocity (VRBC) responses to neuronal activation. These network-wide impairments were fundamentally driven by severely blunted vessel reactivity, including a 28±7% decrease in the magnitude of arteriolar dilations and a 47±7% decrease in the magnitude of arteriolar constrictions. Conclusion: These findings provide a mechanistic foundation for worse clinical outcomes seen in TBI patients with comorbid chronic stress, identifying arteriolar reactivity and long-range capillary coordination as critical therapeutic targets to mitigate secondary injury and improve long-term recovery.

Importance: Conventional repetitive transcranial magnetic stimulation (rTMS) can be ineffective in individuals who have previously failed brain stimulation, ketamine and/or multiple lines of therapies. Modern accelerated rTMS protocols using image-guided targets have not been systematically investigated in these individuals. The goal of this study was to assess the feasibility and efficacy of personalized, connectivity-guided, accelerated intermittent theta-burst stimulation (iTBS) in patients with treatment-resistant depression (TRD) of varying refractoriness. Objective: To assess whether connectivity-guided, accelerated iTBS produces significant reductions in depression severity, and to what extent this benefit extends to ultra treatment-resistant depression (UTRD). Design: This was an open-label feasibility trial of connectivity-guided, accelerated iTBS in patients with TRD. Two distinct groups of participants were recruited from a neurosurgical-psychiatry clinic with UTRD and an interventional psychiatry clinic with TRD. Patients were stratified into a priori treatment-resistance subgroups. Patients received five days of open-label treatment. Outcome measures were collected immediately prior to and after treatment, as well as at 4- and 12-weeks post-treatment. Setting: This trial ( NCT05813093 ) was conducted between November 2023 and July 2025 at Sunnybrook Health Sciences Centre in Toronto, Ontario, Canada. Participants: Patients with major depressive disorder. A total of 96 participants were screened, with 73 meeting eligibility criteria (UTRD=30, TRD=43). One withdrew due to inability to tolerate the baseline MRI, and the other withdrew voluntarily prior to treatment. Intervention: Participants underwent a neuronavigated accelerated iTBS (600 pulses) protocol using personalized left dorsolateral prefrontal cortex (dlPFC) targets derived from functional magnetic resonance imaging (fMRI), comprising eight daily treatments, repeated over five days. Main Outcomes: Primary outcomes were i) change in Hamilton Depression Rating Scale (HAM-D17) from baseline to the end of the fifth day of treatment, and ii) the difference in change in HAM-D17 between UTRD and TRD subgroups. Results: Connectivity-guided fMRI targeting yielded personalized targets clustered around the anterolateral dlPFC. Accelerated iTBS elicited rapid antidepressant effects (ΔHAM-D17 -9.01 [SD 6.06], t = -12.45, p < 0.001) regardless of treatment-resistance group (ΔHAM-D17 -9.64 [SD 5.94] vs -8.10 [SD 6.12], t = -1.05, p = 0.299), which were sustained up to 12 weeks after treatment. Overall response and remission rates at the end of treatment were 40.8% and 16.9%. Self-report scales revealed broad symptomatic relief outside of core depressive symptoms. Conclusions & Relevance: This study demonstrated that fMRI connectivity-guided, accelerated iTBS induces sustained antidepressant effects and broader psychiatric benefits in patients across the spectrum of TRD. In a cohort unlikely to respond to most antidepressant therapies, connectivity-guided, accelerated iTBS offers a safe, well-tolerated option that can achieve benefit, or when ineffective, allow patients to expeditiously proceed with subsequent therapies than conventional rTMS. Trial Registration: This clinical trial was registered at clinicaltrials.gov with NCT05813093 .

Functional hyperemia is a well-established hallmark of healthy brain function, whereby local brain blood flow adjusts in response to a change in the activity of the surrounding neurons. Although functional hyperemia has been extensively studied at the level of both tissue and individual vessels, vascular network-level coordination remains largely unknown. To bridge this gap, we developed a deep learning-based pipeline that uses two-photon fluorescence microscopy images of cerebral microcirculation to enable automated reconstruction and quantification of the geometric changes across the microvascular network, comprising hundreds of interconnected blood vessels, pre and post-activation of the neighboring neurons. The pipeline’s utility was demonstrated in the Thy1-ChR2 optogenetic mouse model, where we observed network-wide vessel radius changes to depend on the photostimulation intensity, with both dilations and constrictions occurring across the cortical depth, at an average of 16.1±14.3 μm (mean ± SD) away from the most proximal neuron for dilations; and at 21.9±14.6 μm away for constrictions. We observed a significant heterogeneity of the vascular radius changes within vessels, with radius adjustment varying by an average of 24 ± 28% of the resting diameter, likely reflecting the heterogeneity of the distribution of contractile cells on the vessel walls. A graph theory-based network analysis revealed that the assortativity of adjacent blood vessel responses rose by 152 ± 65% at 4.3 mW/mm ² of blue photostimulation vs . the control, with a 4% median increase in the efficiency of the capillary networks during this level of blue photostimulation in relation to the baseline. Interrogating individual vessels is thus not sufficient to predict how the blood flow is modulated in the network. Our pipeline, enables tracking of the microvascular network geometry over time, relating caliber adjustments to vessel wall-associated cells’ state, and mapping network-level flow distribution impairments in experimental models of disease.

Study Objectives: We investigate a Mamba-based deep learning approach for sleep staging on signals from ANNE One (Sibel Health, Chicago, IL), a non-intrusive dual-module wireless wearable system measuring chest electrocardiography, triaxial accelerometry, chest temperature, and finger photoplethysmography and finger temperature. Methods: We obtained wearable sensor recordings from 357 adults undergoing concurrent polysomnography at a tertiary care sleep lab. Each polysomnography recording was manually scored, and these annotations served as ground truth labels for training and evaluation of our models. Polysomnography and wearable sensor data were automatically aligned using their electrocardiography channels with manual confirmation by visual inspection. We trained a Mamba-based recurrent neural network architecture on these recordings. Ensembling of model variants with similar architectures was performed. Results: After ensembling, the model attains a 3-class (wake, non-rapid eye movement sleep, rapid eye movement sleep) balanced accuracy of 84.02 per cent, F1 score of 84.23 per cent, Cohen’s κ of 72.89 per cent, and a Matthews correlation coefficient (MCC) score of 73.00 per cent; a 4-class (wake, light NREM [N1/N2], deep NREM [N3], REM) balanced accuracy of 75.30 per cent, F1 score of 74.10 per cent, Cohen’s κ of 61.51 per cent, and MCC score of 61.95 per cent; a 5-class (wake, N1, N2, N3, REM) balanced accuracy of 65.11 per cent, F1 score of 66.15 per cent, Cohen’s κ of 53.23 per cent, MCC score of 54.38 per cent. Conclusions: Our Mamba-based deep learning model can successfully infer major sleep stages from the ANNE One, a wearable system without electroencephalography, and can be applied to data from adults attending a tertiary care sleep clinic.

Enlarged perivascular spaces (PVS) are imaging markers of cerebral small vessel disease (CSVD) that are associated with age, disease phenotypes, and overall health. Quantification of PVS is challenging but necessary to expand an understanding of their role in cerebrovascular pathology. Accurate and automated segmentation of PVS on T ₁ ‐weighted images would be valuable given the widespread use of T ₁ ‐weighted imaging protocols in multisite clinical and research datasets. We introduce segcsvd _PVS , a convolutional neural network (CNN)‐based tool for automated PVS segmentation on T ₁ ‐weighted images. segcsvd _PVS was developed using a novel hierarchical approach that builds on existing tools and incorporates robust training strategies to enhance the accuracy and consistency of PVS segmentation. Performance was evaluated using a comprehensive evaluation strategy that included comparison to existing benchmark methods, ablation‐based validation, accuracy validation against manual ground truth annotations, correlation with age‐related PVS burden as a biological benchmark, and extensive robustness testing. segcsvd _PVS achieved strong object‐level performance for basal ganglia PVS (DSC = 0.78), exhibiting both high sensitivity (SNS = 0.80) and precision (PRC = 0.78). Although voxel‐level precision was lower (PRC = 0.57), manual correction improved this by only ~3%, indicating that the additional voxels reflected primary boundary‐ or extent‐related differences rather than correctable false positive error. For non‐basal ganglia PVS, segcsvd _PVS outperformed benchmark methods, exhibiting higher voxel‐level performance across several metrics (DSC = 0.60, SNS = 0.67, PRC = 0.57, NSD = 0.77), despite overall lower performance relative to basal ganglia PVS. Additionally, the association between age and segmentation‐derived measures of PVS burden was consistently stronger and more reliable for segcsvd _PVS compared to benchmark methods across three cohorts (test6, ADNI, CAHHM), providing further evidence of the accuracy and consistency of its segmentation output. segcsvd _PVS demonstrates robust performance across diverse imaging conditions and improved sensitivity to biologically meaningful associations, supporting its utility as a T ₁ ‐based PVS segmentation tool.