Human Brain Mapping

BackgroundDeep grey matter structures such as the thalamus and basal nuclei are implicated in numerous neurological disorders, yet accurate segmentation of these structures from standard T1-weighted MRI remains challenging due to poor intra-subcortical contrast, long preprocessing pipelines, and fragmented toolsets. MethodsWe introduce THOMASINA a deep learning pipeline for comprehensive subcortical segmentation from standard T1-weighted (T1w) as well as white-matter-nulled (WMn) MRI. The metho...

Deep-learning based super-resolution has shown promise for enhancing the spatial resolution of brain magnetic resonance images, which may help visualize small anatomical structures more clearly. However, when only limited training data are available, it remains uncertain which model assessment method provides the most reliable estimate of out-of-sample performance. In this study, three widely used assessment strategies (three-way holdout, k-fold cross-validation, and nested cross-validation) wer...

AO_SCPLOWBSTRACTC_SCPLOWMagnetic resonance imaging (MRI) is a cornerstone of modern neuroimaging, where accurate segmentation of brain structures and lesions is essential for diagnosis, treatment planning, and clinical research. However, most current foundation models are trained on mixed-organ datasets, while the anatomical structures of the brain differ substantially from those of other organs such as the lungs and kidneys. As a result, these models often struggle to adapt to the distinctive c...

PurposeMELD Graph is a state-of-the-art artificial intelligence (AI) model for automated detection of focal cortical dysplasia (FCD), but its performance remains limited, highlighting the need to investigate which aspects of the pipeline affect its accuracy. MethodsA retrospective failure-mode analysis of the MELD Graph pipeline was performed in 242 subjects, with model predictions and FreeSurfer segmentations reviewed to classify errors as segmentation-associated or algorithm-related. FCD imag...

Resting-state fMRI provides a non-invasive window into large-scale network-level alterations in Alzheimers disease (AD), but the high-dimensional functional connectivity (FC) and multi-site heterogeneity pose challenges to both classification and interpretabil-ity. We propose an explainable deep-learning framework that combines diagnosis-agnostic latent representation learning with a rigorously nested and interpretable classification pipeline to identify reproducible connectivity biomarkers of A...

Neuroimaging faces a reproducibility crisis, where studies on small, heterogeneous datasets produce unreliable brain-wide associations and AI models that fail to generalize. To address this, we introduce GenBrain, a generative foundation model pretrained on approximately 1.2 million 3D scans from over 44,000 individuals across 34 imaging modalities to learn a population prior of brain structure and function. Crucially, GenBrain enables rapid, data-efficient adaptation, allowing any targeted stud...

Large pretrained models developed and shared by actors with privileged access to data and compute have played a central role in the democratisation of deep learning in a range of domains. Here, we contribute to this endeavour in the field of neuroimaging, by compiling a large dataset of structural magnetic resonance imaging scans (n=114,257) and using them to pretrain a multi-task convolutional neural network to predict age, sex, handedness, BMI, fluid intelligence and neuroticism. Subsequent an...

BackgroundSpeech cortical mapping (SCM) conducted with widely available functional MRI (fMRI) can yield divergent results compared to the more commonly used navigated TMS (nTMS). The impact of specific fMRI task paradigms and preprocessing choices on reaching similarity with nTMS has not been explored before. ObjectiveTo test how the fMRI experimental task and spatial smoothing of the data compare with nTMS-based results, to subsequently increase the reliability of object naming fMRI for SCM. ...

We present PANDORA (Population Archive of Neuroimaging Data Organized for Rapid Analysis), a huge brain imaging data archive and analysis resource for UK Biobank neuroimaging data. PANDORA UKBv1 contains 81,939 subjects voxel-level images, created by the core UKB brain image processing pipeline. PANDORA also includes highly efficient supervoxel versions of the data - much smaller and faster to work with than the full voxelwise representation while losing virtually no signal or spatial detail and...

The brain age gap (BAG) -- the difference between chronological and neuroimaging-based predicted brain age -- has emerged as a sensitive biomarker of brain health. A higher BAG, reflecting an older-appearing brain, has been linked to cognitive decline, neurodegenerative disease, and mental disorders. Whether such apparent aging can be mitigated by targeted interventions remains unclear. Oestradiol (E2), a sex hormone fluctuating across the menstrual cycle and known for its neuromodulatory and co...

BackgroundMotor threshold (MT) estimation is fundamental to transcranial magnetic stimulation (TMS), guiding individualized stimulation intensity in research and therapy. Conventional methods such as the 5-out-of-10 rule require many stimuli, while adaptive approaches like Parameter Estimation by Sequential Testing (PEST) improve efficiency but can exhibit poor convergence under certain conditions. ObjectiveThis study introduces the Bayesian Uncertainty Dynamic Algorithm for Parameter Estimatio...

Lesion network mapping (LNM) links focal brain lesions to distributed neural circuits by projecting lesion locations through a normative functional connectome. van den Heuvel and colleagues recently showed how commonly used LNM procedures generate maps that converge on nonspecific, low-dimensional properties of the connectome, introducing a bias. Consequently, many published maps of different conditions appear strikingly similar. Here, we offer an alternative approach that does highlight distinc...

Magnetic Resonance Spectroscopy Imaging (MRSI) offers spatially-resolved, neurometabolic information, acquired non-invasively at whole-brain scales from human subjects. Analysis of MRSI however, is extremely challenging. The metabolic information is highly convolved, and sparsely distributed across millions of spatial-spectral datapoints, allowing for little direct human interpretation. Conversely, the overall low signal-to-noise with high-intensity artifacts can confound unsupervised machine le...

Autism spectrum disorder (ASD) is increasingly conceptualized as a disorder of large-scale functional brain network organization rather than isolated regional abnormalities. Graph-theoretical analysis provides a principled framework for characterizing such distributed network reconfiguration. Here, we investigated global, nodal, and system-level functional network topology in ASD using a large, multi-site resting-state fMRI dataset. Resting-state fMRI data from 996 participants (428 ASD, 568 hea...

BackgroundAccurate forecasting of Alzheimers Disease (AD) progression is critical for personalized patient management and clinical trial stratification. However, current predictive models often struggle to effectively integrate high-dimensional neuroimaging with longitudinal clinical data. We introduce AD-LLaVA-3D, a novel multimodal framework designed to bridge this gap by adapting large vision-language models for volumetric and temporal forecasting. MethodsWe leveraged the LLaVA-NeXT-Video ar...

Task-based functional magnetic resonance imaging (fMRI) examines how the brain dynamically responds to cognitive and perceptual demands, offering complementary insight beyond traditional activation-based analyses in schizophrenia. Prior task-based fMRI studies have identified reduced functional connectivity within auditory and associated cortical areas. In this study, we investigated task-evoked functional connectivity and brain state dynamics in 25 healthy controls, 23 patients with schizophren...

Heart-brain interactions, including cardiac-induced brain pulsatility, are thought to support brain homeostasis, yet their alteration with aging and neurodegeneration remains poorly understood. Here, we used three-dimensional quantitative amplified MRI (q-aMRI) to visualize and quantify cardiac-induced pulsatile brain motion across healthy individuals and those on the Alzheimers and Lewy body disease spectra. Expert evaluations revealed consistent distinctions between normal and abnormal motion ...

The effect of biological aging on brain structure is widespread and apparent. However, little is understood regarding which regions exhibit similarities in vulnerability, and what biological processes drive regional patterns of senescence-associated atrophy. Here, we investigated whether associations between age and brain structure exhibit distinct patterns of regional vulnerability, and if so, whether they are related to patterns of cerebral physiology which also show age-related decline. Using...

BackgroundMild cognitive impairment (MCI) precedes Alzheimers disease (AD) in [~]40% of cases, with early language deficits distinguishing converters. This study develops a DTI radiomics model from language network gray matter to predict MCI to AD conversion and identify preclinical biomarkers. MethodsThis retrospective case-control study analyzed diffusion tensor imaging (DTI) data from 97 individuals with MCI (29 converters, 68 non-converters) from the Alzheimers Disease Neuroimaging Initiati...

IntroductionThe balance between excitatory and inhibitory (E/I) neural processes is a fundamental principle of brain function, and its disruption has been implicated in the pathophysiology of multiple sclerosis (MS). In vivo assessment of E/I balance has traditionally relied on electrophysiological measures, and despite the abundance of fMRI data on MS, no fMRI-based technique has so far been presented to measure E/I balance in MS. MethodsRecently, a novel MRI-based method has been introduced t...