No published license or data use terms

Datasets under this license

• ACSID (FOR2974)

  Task fMRI from German studies of gaming, pornography, buying-shopping and social-network use disorders, run to find the cognitive and neural mechanisms behind addictive internet use. Imaging covers a few subprojects, such as a conditioning study in 83 men. Not public, data on request.

• MGH antisaccade error processing (fMRI and MEG/EEG)

  Brain MRI with antisaccade task fMRI of healthy adults and people with schizophrenia, OCD or autism, who also did the same task during simultaneous MEG and EEG, collected to test how DRD4 and MTHFR variants relate to two error markers. On request only.

• MIND diet trial brain MRI (AGMP_MIND)

  Brain MRI from a three-year randomized trial testing the MIND diet against a control diet in cognitively unimpaired, overweight older adults with a family history of dementia. 267 people scanned at baseline, 201 again at year 3. The images are not publicly released.

• ALLOP1

  Brain MRI from an early-phase trial of intravenous allopregnanolone versus placebo in people with MCI or mild Alzheimer's disease, testing the neurosteroid as a regenerative treatment. 24 participants scanned before and after 12 weeks at one site, held in the LONI IDA.

• ANDES schizophrenia MRI (six Latin American cities)

  Structural brain MRI of patients with schizophrenia and healthy controls from six Latin American cities, pooled to study how neighbourhood poverty and violence relate to grey matter volume. 334 patients and 262 controls, with neighbourhood income and homicide rates. On request only.

• CONNECT/ARCHI

  Multimodal brain MRI of healthy young adults (diffusion, anatomical, relaxometry and resting-state fMRI), acquired to build probabilistic atlases of human white matter and cortical connectivity. Images are not released; derived connectivity data are on EBRAINS.

• Asan Medical Center SWI Microbleed Set (ARIA-H)

  Susceptibility-weighted brain MRI labelled for microbleeds, used to train a detector that grades ARIA-H severity during anti-amyloid therapy. 565 scans from Asan Medical Center, Seoul, 429 with microbleeds. Described in a 2025 conference abstract, not publicly released.

• Cerebellar ataxia FDG PET (Asan)

  Retrospective brain FDG PET of patients with cerebellar-type multiple system atrophy and spinocerebellar ataxia types 2 and 6, compared with crossed cerebellar diaschisis and healthy controls to map cerebellar subregional metabolism. 171 subjects from Seoul. Images on request.

• Austrian Stroke Prevention Family Study (ASPS-Fam)

  Brain MRI, including diffusion and R2* iron mapping, and cardiac CT from stroke- and dementia-free adults and their relatives in Graz, collected to study how vascular risk factors and family history affect brain structure and vessels. Images on request only.

• AURORA (Leiden D-CAA cohort)

  Yearly brain MRI, CSF and cognitive tests of presymptomatic and symptomatic carriers of hereditary cerebral amyloid angiopathy (Dutch type), followed to learn the order in which biomarkers change. 120 people in Leiden, scanned at 3 T and 7 T. Data on request from the authors.

• Brazilian High Risk Cohort (BHRC)

  Longitudinal brain MRI with psychiatric, cognitive and genetic data from Brazilian school children enriched for risk of mental disorders, followed to study how psychiatric problems develop. 1,321 young people scanned in up to four waves from 2010 to 2025.

• BIODEP

  Brain MRI and TSPO PET/MR of adults with major depression and healthy controls, grouped by blood CRP, to test whether inflammation marks a subgroup of depression. 143 adults aged 25 to 50 with MRI, 76 with PET. No published data sharing route.

• Emotion Regulation fMRI in Borderline and Cluster-C PD (Participants' MRI scans, which are not shared)

  Task fMRI of women with borderline personality disorder, women with cluster-C personality disorder and nonpatient controls while they viewed or regulated responses to emotional pictures. Scans are not public; group-level BrainVoyager statistics are open under CC0.

• Brain Before Stroke (BBS)

  Early brain MRI after a first ischemic stroke with 1-year outcomes, collected to study how imaging markers relate to recovery. 428 adults from a prospective cohort in Bordeaux, France, scanned 24 to 72 hours after onset. Shared on request to the principal investigator.

• Brainlinks Adolescent Trust fMRI

  Task and resting-state fMRI with T1-weighted MRI from adolescents and young adults playing a trust game with a friend, a community member and an unknown peer, collected to study how trust and its brain basis differ by partner and age. Shared on request.

• Breast lesion DCE-MRI radiomics (Naples) (MR images and lesion segmentations.)

  Breast dynamic contrast-enhanced MRI of 85 women with histologically proven lesions, used to tell benign from malignant lesions with texture, shape and perfusion features. Only per-lesion shape features are open; the images are not shared.

• Cambridge Foetal Testosterone (CAM-FT) study, MRI cohort

  Structural brain MRI of typically developing boys whose testosterone was measured in amniotic fluid before birth, used to relate prenatal testosterone to regional grey matter volume. 28 boys aged 8 to 11 from Cambridgeshire. Not publicly shared.

• D-dimer and Cerebral Infarction in ICU Cancer Patients

  Brain MRI of adults with cancer in a medical oncology intensive care unit, scanned for suspected acute ischemic stroke, collected to study stroke patterns and D-dimer levels. Images are not shared; a per-patient clinical table is public.

• Cao 2026 In-house cSS Test Set

  Brain MRI of patients with cortical superficial siderosis, a marker of cerebral amyloid angiopathy, with manual lesion outlines checked by a neuroradiologist, used to test segmentation trained on synthetic data. 10 patients with SWI or T2*-weighted scans; not released.

• Cardiac Atlas Project CHD Atlas

  Cardiac MRI, clinical data and heart shape models of children and adults with congenital heart disease, mostly tetralogy of Fallot, collected as a reference population to compare a patient's ventricles against. 1,183 people from San Diego and Auckland, shared under a data use agreement.

• Cerebral Perfusion and Cognitive Decline in Type 2 Diabetes

  Brain MRI with arterial spin labeling perfusion of older adults with type 2 diabetes and healthy controls, collected to study how diabetes affects cerebral blood flow regulation and cognition. Images are not shared; derived MRI measures and physiological recordings are public.

• Cerebromicrovascular Disease in Elderly with Diabetes

  Brain MRI with arterial spin labeling perfusion, FLAIR and diffusion imaging of older adults with type 2 diabetes and controls, at baseline and two-year follow-up, to study diabetic effects on cerebral vasoregulation. Images are not shared; derived MRI measures are public.

• CHBMP

  Brain MRI, high-density resting-state EEG, cognitive tests and demographics of healthy Cuban adults, collected to build normative references for EEG and its brain sources. 282 people aged 18 to 68, 203 with structural and diffusion MRI, shared in BIDS on Synapse.

• Bone chondrosarcoma MRI radiomics (Milan)

  Unenhanced T1- and T2-weighted MRI of cartilaginous bone tumours, collected to grade low-grade atypical cartilaginous tumours against higher-grade chondrosarcoma with radiomics and machine learning. Files are restricted on Zenodo and not openly available.

• Clarity AD (lecanemab phase 3 trial imaging)

  Safety brain MRI from the lecanemab phase 3 trial in early Alzheimer's disease, read centrally for amyloid-related imaging abnormalities (ARIA) during treatment. 1,795 people scanned at 1.5 T or 3 T. The images are not shared outside the sponsor.

• CMR with device susceptibility artifacts (Penso 2022)

  Cardiac cine MRI of patients with and without pacemakers or implantable defibrillators, collected to train and test automatic segmentation of both ventricles and the left ventricular myocardium in images degraded by metal susceptibility artifacts. Two Italian centers, expert contours. Access restricted.

• CMU SC and BOLD fMRI (T1-weighted, diffusion spectrum and resting-state fMRI images.)

  Diffusion spectrum, resting-state fMRI and T1-weighted brain MRI of healthy young adults, used to study hierarchical community structure in structural and functional brain networks. Only streamline count matrices and regional BOLD time series are public; the images are not.

• CMUH SWAN Microbleed Set

  Clinical susceptibility-weighted brain MRI with and without cerebral microbleeds, with voxel-wise microbleed masks, used to test whether detectors trained on synthetic microbleeds work on real scans. 205 exams from a hospital in Taiwan, 104 with microbleeds. Shared only through collaboration.

• COVORTS (f)MRI sub-study

  Structural and odour-task functional brain MRI of adults with persistent smell loss or parosmia after COVID-19, collected to study olfactory bulb volume, grey matter and odour-evoked brain activity. Scans are not shared; the data record holds questionnaire and smell-test data.

• Colorectal liver metastasis DWI-DKI RAS (Naples) (MR images and lesion segmentations.)

  Liver MRI with multi-b-value diffusion of 52 patients with resected colorectal liver metastases, used to predict RAS mutation from diffusion, IVIM and kurtosis parameters. Only per-lesion parameter values are open; the images are not shared.

• Dartmouth Backward Masking Pulvinar VBM

  T1-weighted brain MRI of healthy young adults pooled from two Dartmouth backward masking fMRI studies (64 subjects), used to compare grey matter volume between people aware and unaware of masked faces. Raw images are not shared; per-subject VBM maps are public.

• DeepMultiBrain Bordeaux 3T multimodal MRI

  Brain MRI of young healthy volunteers at Bordeaux Hospital on a Canon 3T scanner: 0.6 mm T1w and T2w plus 0.4 mm white-matter-nulled scans for 55 subjects, used to train WMn synthesis for the holiAtlas. Private, not shared.

• DERIVATE-ICM registry

  Cardiac MRI of patients with ischemic cardiomyopathy and chronic heart failure, collected to test whether MRI improves the choice of who should receive a preventive implantable cardioverter-defibrillator compared with echocardiography. Multicenter, with arrhythmic outcomes over follow-up. Access restricted.

• Depression Biotypes Multisite rsfMRI (Drysdale 2017) (Patient and control scans from the investigator-held sites (Cornell 1, Cornell 2, Emory 1, Stanford 1, Stanford 2, Toronto and Harvard).)

  Resting-state fMRI and structural brain MRI of adults with major depression and healthy controls, pooled to define four connectivity subtypes of depression. 458 patients and 730 controls from 17 sites. Patient sites share data at the investigators' discretion.

• Edinburgh PVS 1.5T vs 3T Pilot

  Brain MRI of healthy adult volunteers scanned back to back at 1.5T and 3T, collected to test whether perivascular space measurements can be harmonised across field strengths. Images are not shared; PVS scores, measurements and harmonisation scripts are public.

• EMERGE and ENGAGE (aducanumab phase 3 trial imaging)

  Safety brain MRI and amyloid PET from two aducanumab phase 3 trials in early Alzheimer's disease, read for amyloid-related imaging abnormalities (ARIA) during treatment. 3,285 people across both trials. The images are not shared outside the sponsor.

• EMIT

  Multiplexed immunofluorescence microscopy of a human tissue microarray with cores from 34 types of cancer, non-neoplastic disease and normal tissue, released as reference data for developing and benchmarking tissue image processing methods such as cell segmentation.

• EXALT exenatide trial in alcohol use disorder (Brain imaging (fMRI with structural MRI, and DAT SPECT), shared on request only.)

  Brain fMRI, structural MRI and dopamine transporter SPECT from a Copenhagen trial of weekly exenatide versus placebo in alcohol use disorder, scanned at baseline and week 26, with healthy controls. Imaging on request; clinical tables are open under CC BY 4.0.

• FAB4V Stroke Diffusion MRI

  Diffusion MRI and FLAIR of first-ever ischemic stroke patients with manual lesion masks, used to test how well healthy-control tractography predicts white matter disconnection. Images are not shared; a lesion overlap map and statistical maps are public.

• Failed total hip arthroplasty MRI (Milan) (MR images.)

  Pelvis MRI of patients with a failed total hip arthroplasty, read for periprosthetic findings and regional lymph nodes to tell infected from aseptic failure, using a metal artifact reduction protocol at 1.5 T. Only a spreadsheet of readings is public; images are not shared.

• Children and Adults Reading Fairy Tales fMRI

  Task fMRI of children and adults reading Chinese passages from four Hans Christian Andersen fairy tales, acquired at Peking University and preprocessed with SPM12, deposited to study the brain during naturalistic story reading, with restricted access through Science Data Bank.

• Feedback and Learning Pace Memory fMRI (Individual fMRI scans, which are not shared)

  Task fMRI study of participants learning items at their own pace or at a fixed pace, with or without feedback and with reward or punishment feedback. Only group-level SPM results and behavioral tables are public; individual scans are not shared.

• Florbetaben Phase IIb Trial (diagnostic confidence add-on)

  Amyloid PET with florbetaben in people with probable Alzheimer's disease and cognitively unimpaired controls, acquired in a phase IIb diagnostic trial and used to study how scan results change referring physicians' diagnostic confidence. 272 people. The images are not shared.

• Prague Functional Movement Disorder Resting-State fMRI (Pre-processed functional MRI, shared on request)

  Resting-state BOLD fMRI of adults with functional movement disorder and healthy controls, used to relate brain network centrality to motor symptom severity. Images are shared on request; per-subject connectivity maps are public.

• FOCAS (Leiden sporadic CAA cohort)

  Yearly brain MRI, CSF and clinical follow-up of people with sporadic cerebral amyloid angiopathy, collected to study how its MRI markers progress over time. 126 participants in Leiden, scanned at 3 T and 7 T, with siderosis masks for a small subset. Data on request from the authors.

• Frontostriatal Reward Learning fMRI (IPCAS)

  Task fMRI of healthy participants performing a probabilistic instrumental learning task with reward, punishment and neutral outcomes (80 scanned), deposited with EEG and brain stimulation study data from the Institute of Psychology, Chinese Academy of Sciences, with restricted access through Science Data Bank.

• Generation R Study (neuroimaging)

  Repeated brain MRI of children from a population-based birth cohort in Rotterdam, collected to study how prenatal and early-life exposures, genetics and behaviour relate to brain development. Structural, diffusion and resting-state functional MRI. Access through collaboration requests.

• GROUP Maastricht MRI

  Structural and diffusion brain MRI of patients with non-affective psychosis, their unaffected siblings and healthy controls, from a Dutch study of genetic risk and outcome of psychosis. 89 patients, 95 siblings and 87 controls from the Maastricht site. Shared only on request.

• GTU Lumbar MRI Dataset

  Sagittal lumbar spine MRI slices with expert-marked centre points of each vertebra and disc, a small benchmark for locating and labelling spine landmarks. 80 subjects, 64 with disc or vertebra pathologies, with T1- and T2-weighted slices.

• GUSTO

  Brain MRI shortly after birth and repeatedly through childhood from a Singapore birth cohort of Chinese, Malay and Indian families, which studies how early-life factors shape metabolic health, neurodevelopment and mental health. Families recruited in pregnancy in 2009 and 2010. Shared through collaboration.

• Fast thoracic MRI versus chest x-ray (San Donato)

  Chest MRI with a fast single-shot turbo spin-echo sequence and chest x-ray from the same patients, collected to compare how often each detects lung nodules, consolidation, pleural effusion, cardiac enlargement and aortic dilation. Retrospective, single country. Access restricted.

• Optic Radiation HARDI Tractography (Hospital Clinic)

  Brain MRI with two single-shell HARDI diffusion protocols and 3D T1-weighted imaging of healthy volunteers and multiple sclerosis patients, collected to reconstruct the optic radiation. Images are not shared; only a group tractography template is public.

• Ribeirão Preto Multiple Sclerosis Quantitative MRI Cohort

  Brain MRI of multiple sclerosis patients and neurologically normal controls, used to classify patients from regional quantitative measures (diffusion, T2 relaxometry, magnetization transfer, morphometry). Images are not shared; only the paper's figures and result tables are public.

• HDNI (TRACK-HD)

  Brain MRI from people with premanifest and early Huntington's disease and matched controls, collected in the TRACK-HD natural history study to find measures that track disease progression for clinical trials. Images were archived at LONI; curated copies are shared by CHDI under a data use agreement.

• San Raffaele MS Functional Network and Gene Expression

  Resting-state functional and structural brain MRI of people with multiple sclerosis and healthy controls, used to relate network centrality changes to regional brain gene expression. Images are not shared; a table of clinical and volumetric measures is public.

• Diffusion MRI of Myelopathy in Adrenoleukodystrophy

  Brain and cervical spinal cord diffusion MRI from men with adrenoleukodystrophy and controls, collected to test diffusion metrics as outcome measures for myelopathy progression over two years. Images are not shared; derived tables are public.

• Methionine PET for glioblastoma radiotherapy (Humanitas)

  Postoperative amino acid PET/CT and brain MRI of adults with newly diagnosed glioblastoma treated with hypofractionated radiotherapy and temozolomide in a phase II trial, used to relate methionine uptake to residual tumour, target volumes and survival. 93 patients. Data on request.

• Methionine PET in lower-grade glioma (Humanitas)

  Amino acid PET/CT and brain MRI before surgery in adults with newly diagnosed grade 2 or 3 glioma, used to relate [11C]methionine uptake to tumour type, tumour extent and progression-free survival. 153 patients from one Italian centre. Data on request.

• Tractography of the Corpus Callosum in Huntington's Disease

  Brain MRI with diffusion tensor imaging of presymptomatic and manifest Huntington's disease gene carriers and healthy controls, collected to map damage along the corpus callosum tracts. Images are not shared; only summary tract tables are public.

• IMAP+ (Caen)

  Brain MRI, FDG and amyloid PET, biomarkers and cognition of healthy adults and people with subjective decline, MCI or Alzheimer's disease, collected to find early markers of the disease and follow their change. 242 people from one centre in Caen, France. Shared on request.

• Resting-State fMRI Clustering Cohort (Karolinska) (Participants' resting-state fMRI, which is not shared)

  Brain BOLD fMRI at rest from healthy adults aged 21 to 84, collected to test voxel-wise hierarchical clustering of whole-brain functional connectivity networks. The scans are not public; only the paper and its supplementary figures are.

• KNE96 Korean Normal Elderly template cohort

  Structural brain MRI of cognitively normal older Korean adults, used to build elderly brain templates for spatial normalization in Asian populations. 144 people aged 60 or older; only the derived templates were released, not the scans.

• LCBC

  Repeated brain MRI and cognitive tests of healthy people from early childhood to old age, used to model lifespan brain trajectories and early markers of Alzheimer's disease. 4,352 scans of 2,017 participants aged 4 to 93 from Oslo. Not public; on request to the principal investigator.

• Leipzig Heart Failure Resting-State fMRI (Pre-processed functional MRI, shared on request)

  Resting-state BOLD fMRI of cardiac patients with and without coronary artery disease and heart failure, used to relate brain network centrality to heart function and cognition. Images are shared on request; per-subject connectivity maps are public.

• LGE gadobutrol dose CMR in chronic MI (San Donato)

  Cardiac MRI with cine and late gadolinium enhancement of 79 patients with chronic transmural myocardial infarction, scanned after 0.10, 0.15 or 0.20 mmol/kg gadobutrol, with scar segmentation and SNR/CNR measures. One Italian centre. Access on request.

• LINCHPIN CT and pathology set (CT images)

  Head CT ratings, APOE genotype and autopsy grading of cerebral amyloid angiopathy from adults who died after intracerebral haemorrhage, used to derive the Edinburgh criteria for CAA-related bleeds. 110 adults from Lothian, Scotland. The ratings are open; the CT images are not shared.

• Brain qMRI after Liver Transplantation (Hannover)

  Quantitative brain MRI of adults about ten years after liver transplantation on different calcineurin inhibitor doses and matched healthy controls, collected to study long-term effects of immunosuppression on brain tissue. Images are not shared; only group-level result tables are public.

• Endotoxin Sickness Insula Resting-State fMRI

  Resting-state brain fMRI from a randomized, double-blind trial in which 48 healthy volunteers were scanned about 2 h 45 min after low-dose endotoxin (LPS) or placebo, to study insula connectivity during sickness. Raw images are not shared; group t maps are on NeuroVault.

• Marburg Co-speech Gesture IFG fMRI

  Task fMRI of adults judging whether speech and co-speech gestures in short videos match, collected to compare inferior frontal gyrus activation and laterality for metaphoric, iconic and unrelated gestures. Raw images are not shared; per-subject contrast maps are public.

• MDD Belief-Updating fMRI (Ultimatum Game)

  Task fMRI study data from unmedicated adults with major depressive disorder and healthy controls playing an ultimatum game, deposited to study how the insular-cingulate network updates beliefs from prediction errors, with restricted access through Science Data Bank.

• MDD brain-derived subtypes (San Raffaele) (MR images.)

  Structural and diffusion brain MRI features of 102 inpatients with major depressive disorder, used to cluster patients into neuroimaging subtypes linked to treatment resistance and inflammation. Only regional feature tables and clinical variables are open; the images are not shared.

• 11C-methionine PET and DWI in operated gliomas (Humanitas)

  Preoperative amino acid PET and diffusion-weighted brain MRI of glioma patients from one Milan-area centre, used to compare methionine uptake and diffusion for grading, molecular profiling and progression-free survival against histopathology. 124 patients. Access restricted.

• MGH CAA cohort, cSS evolution set

  Baseline and follow-up brain MRI of patients with probable cerebral amyloid angiopathy and no prior brain bleed, rated for siderosis progression and new microbleeds, used to show that siderosis progression predicts a first bleed. 118 patients from Massachusetts General Hospital. Not public.

• Motor neuron disease connectome (Milan, Turin, Naples)

  Structural, diffusion and resting-state functional brain MRI of patients with amyotrophic lateral sclerosis, primary lateral sclerosis or progressive muscular atrophy and healthy controls, collected to map brain network changes in motor neuron diseases. Images shared on request.

• Medication overuse headache triple network fMRI (Besta)

  Resting-state fMRI and structural brain MRI of adults with chronic migraine and medication overuse headache and of healthy controls, collected to study default mode, salience and executive network connectivity and its links to drug overuse, allodynia and mood. Access restricted.

• MOSAIC

  Digitized H&E slides, spatial transcriptomics, single-cell and bulk RNA and exome sequencing with clinical data from cancer patients, built to find cancer subtypes, drug targets and biomarkers. 2,725 patients across 11 cancer types, through Owkin's paid K Pro platform.

• Mount Sinai COVID-19 Neuroimaging, Cancer vs Non-Cancer

  Clinical brain MRI of adults hospitalized with COVID-19 and neurological symptoms at Mount Sinai, New York, in four groups of 20 by cancer and ventilatory support status, read by neuroradiologists for infarcts, hemorrhages and hyperintensities. Images are not shared.

• NAPLS 2

  Longitudinal brain MRI with clinical, cognitive and blood data from young people at clinical high risk for psychosis and healthy controls, followed for two years to find what predicts and drives conversion to psychosis. Eight sites in the United States and Canada.

• NEFROCOG Pre-dialysis CKD Brain MRI

  Brain MRI of adults with chronic kidney disease not yet on dialysis, collected to test whether mild cognitive impairment relates to kidney function, inflammatory markers and visible brain changes. Images are not shared; only the article figures and tables are public.

• Neonatal Encephalopathy MRI, Kozhikode (South India)

  Brain MRI, diffusion tensor imaging and thalamic MR spectroscopy of term newborns with encephalopathy after perinatal asphyxia, collected to describe the pattern of brain injury and its link to outcome at age 3.5 years in a low- and middle-income setting. Images are not shared.

• NEUROLINGUA (ds007111) (Original T1-weighted images, shared by BCBL on request only.)

  Brain fMRI of healthy adults performing a language localizer in Spanish, Basque or English, pooled to study how the language comprehension network varies with bilingual and multilingual profiles. Includes structural MRI derivatives and sociolinguistic and cognitive measures.

• NF1 MPNST Imaging Predictors (NTAP)

  MRI and clinical data of people with neurofibromatosis type 1 (NF1), gathered from several centers to find radiomic features that tell malignant peripheral nerve sheath tumors from benign plexiform neurofibromas. Held on the NF Data Portal under embargo.

• TLE structural networks and tonic-clonic seizures (NHNN) (MR images.)

  Diffusion MRI structural brain networks of adults with drug-resistant temporal lobe epilepsy and healthy controls, built to test whether seizures that spread to both hemispheres come with wider network abnormality. Only the connectivity matrices are open; the images are not shared.

• NIMH Child Psychiatry Branch Brain Development Study

  Structural brain MRI of children and young adults scanned about every two years since 1989, about half typically developing and half with ADHD, autism or childhood-onset schizophrenia, to chart brain development. Over 6,000 scans of over 2,000 youth. No public release of the images was found.

• Working memory in schizophrenia and siblings (ds000115)

  Brain MRI and N-back working memory fMRI of people with schizophrenia, their siblings, healthy controls and the controls' siblings, collected to study how brain network connectivity changes with working memory load and with familial risk for schizophrenia.

• Harm avoidance and gray matter volume (ds000202)

  Brain T1-weighted MRI of young healthy women with Temperament and Character Inventory scores, collected to test how VBM and ROI processing choices change the link between harm avoidance and gray matter volume. 95 subjects, 3T Philips, in BIDS.

• OXT_AVP_PLC_fmridata

  Task fMRI of healthy young adults given intranasal oxytocin, vasopressin or placebo while rating empathy for social pain in video clips, collected to compare how the two hormones change empathy and its brain basis. Raw images are not shared; per-subject contrast maps are public.

• OXVASC T2*-GRE Microbleed Set

  Brain MRI of patients after minor stroke or TIA with manual microbleed segmentations, used to train and test a deep learning microbleed detector. 74 patients from a population-based Oxford study, 366 microbleeds in 36 of them. Shared on request to the principal investigator.

• PAD inter-leg symmetry CE-MRA (San Donato)

  Contrast-enhanced MR angiography of the lower limbs from aortic bifurcation to tibial arteries in 82 patients with known or suspected peripheral artery disease, with per-artery stenosis scores for both legs. One Italian centre. Access on request.

• Neonatal brain MRI and cranial ultrasound (Passo Fundo) (MRI and ultrasound examinations of the 51 neonates.)

  Brain MRI and transfontanellar ultrasound of mostly preterm newborns at a Brazilian tertiary hospital for high-risk pregnancy, read for germinal matrix hemorrhage, hydrocephalus and leukomalacia to judge ultrasound against MRI. 51 neonates. Images not shared.

• Prague Personality Resting-State fMRI

  Resting-state brain fMRI and structural MRI of healthy adults with Big Five personality scores, collected in Prague to replicate a reported link between personality and seed-based functional connectivity. Raw images are not shared; per-subject seed beta maps are public.

• Precision Imaging Parkinson's Disease (PIPD)

  Structural and resting-state functional brain MRI of people with Parkinson's disease and healthy controls, collected to map how subcortical targets of Parkinson's treatments connect to the motor cortex. Five resting-state runs per person; files withdrawn from public access, available on request.

• [11C]PK11195 PET-MRI in mitochondrial disease

  Brain PET-MRI of the mitochondrial translocator protein (TSPO) in patients with genetically confirmed mitochondrial disease and matched healthy controls, to test TSPO radioligand binding as an in vivo biomarker of mitochondrial pathology. Raw data on request.

• Visual cortex in macular degeneration (Prins et al. 2016)

  T1-weighted brain MRI of people with juvenile or age-related macular degeneration and age-matched healthy controls, shared to study cortical thickness, surface area and grey matter volume of the visual cortex after central vision loss. Scanned at three European sites.

• QIN-PROSTATE

  Multiparametric prostate MRI of prostate cancer patients, shared to develop quantitative methods such as pharmacokinetic models of contrast uptake. 22 patients from Brigham and Women's Hospital. TCIA retired the collection in April 2026.

• Radboudumc TBI SWI Microbleed Set

  Susceptibility-weighted brain MRI of patients with moderate to severe traumatic brain injury and controls, with microbleed segmentations, used to develop detectors of trauma-related microbleeds. 81 studies of 45 patients and 18 controls from one Dutch centre. Not publicly released.

• Tehran Raven Task fMRI in Recent Suicide Attempters (T1-weighted and functional MRI, which are not shared)

  Task fMRI during a Raven matrices reasoning task in adults with major depressive disorder who recently attempted suicide, depressed patients without an attempt and healthy controls. Scans are not public; group-wise ALFF maps and clinical scores are open under CC0.

• Robert Debré Hospital autism MRI

  Structural and resting-state brain MRI of people with autism and controls from a Paris hospital, used as the unseen test site of a challenge to predict autism diagnosis from MRI. 247 people, 49 with autism. Not released.

• Cerebellum in Parkinson's disease FDG-PET (Rennes)

  Resting-state FDG-PET brain scans of patients with Parkinson's disease, analysed to relate cerebellar glucose metabolism to motor, cognitive and psychiatric scores. The images are not public; the data deposit holds only the supplementary tables and figures.

• NN104 RHAPSODY

  Follow-up brain MRI from a phase two trial in acute ischemic stroke, read centrally for bleeding into the brain while testing an engineered activated protein C (3K3A-APC) given after clot-dissolving or clot-removal treatment. GRE or SWI scans at days 7, 30 and 90.

• Rotterdam Study (Rotterdam Scan Study)

  Repeated brain MRI of middle-aged and older adults from a Dutch population-based cohort, collected to study causes and early signs of dementia, stroke and small vessel disease, with long clinical follow-up. Also CT and MRI of arteries and knees. On request.

• Saguenay Youth Study (SYS)

  Brain and abdominal MRI, genotypes and cardiometabolic and mental health data of adolescents and their parents from a French-Canadian founder population, collected to study how genes and maternal smoking in pregnancy shape the brain and heart health. 1,029 adolescents and 664 parents. On request.

• Porto Alegre PAH Cardiac MRI Cohort

  Cardiac MRI of patients with pulmonary arterial hypertension, used to compare right atrial area and right ventricular ejection fraction with catheterization, echocardiography, walk test and NT-proBNP markers. Images are not shared; only the paper's figures and tables are public.

• Scottish Bipolar Family Study fMRI (Whalley 2013)

  Task fMRI and structural brain MRI of young unaffected relatives of people with bipolar disorder and matched controls, followed for about two years to test whether baseline brain activation predicts later major depression.

• SG-FCCM

  Susceptibility-weighted brain MRI of patients with familial cerebral cavernous malformations, with lesion masks, used to build a pipeline that segments, counts and compares the lesions between exams. 101 scans of 73 patients. Code and weights are public; the data are not.

• St George's COPD and CAD brain cohort (Brain MR images and retinal photographs.)

  Brain MRI (T1, FLAIR, diffusion tensor, arterial spin labelling) and retinal photographs of smokers with chronic obstructive pulmonary disease or coronary artery disease, collected to find vascular, respiratory and inflammatory causes of brain damage. Only derived measures are open.

• SHIVA cohort

  Brain MRI and blood biomarkers of adults over 60 with extensive or minimal white matter lesions, collected to find markers of cerebral small vessel disease and predict cognitive decline. Ongoing French cohort, 150 enrolled by March 2024. Not public.

• Social Media Addiction Time Perception fMRI

  Task fMRI of adults with social media addiction and healthy controls judging time intervals, collected to study how time perception and its brain basis differ in this group. Raw images are not shared; per-subject t maps are public.

• Brain Aging and Speech Perception in Noise (Participants' MRI scans, which are not shared)

  Task fMRI and structural and diffusion MRI of healthy younger and older adults, collected to study how age, background noise and talker variability affect syllable discrimination. Scans are not public; group surface maps and task stimuli are open under CC BY-NC 4.0.

• SpineWeb Dataset 1

  Paired lumbar spine CT and MRI of the same subjects, released for a spine imaging workshop without labels. 30 pairs, shared after a request form describing the research.

• SpineWeb Dataset 10 (multi-modality vertebra recognition)

  Spine MRI and CT with the center and orientation of every vertebra marked by hand, released as reference data for vertebra recognition. 20 subjects, shared after a request form.

• SpineWeb Dataset 11 (anisotropic lumbar MRI)

  Routine lumbar spine MRI with thick slices and manual vertebral body segmentations, released with code for vertebra detection and segmentation on real clinical scans. 17 volumes from several hospitals.

• SpineWeb Dataset 6 (IVD Dixon MRI)

  Lower spine MRI with manual masks of the intervertebral discs, for locating and segmenting the discs. 8 patients with four aligned Dixon images each. Shared on request.

• SpineWeb Dataset 7 (IVD T2 TSE MRI)

  Lower spine MRI with manual masks of the intervertebral discs, for locating and segmenting the discs. 15 patients with 3D T2-weighted scans. Shared on request.

• STRIvE eating disorders MRI

  Brain MRI, task fMRI and spectroscopy of women with binge-eating disorders and controls under acute stress, collected to study how stress affects self-control. 85 women from Cambridge, UK, with anorexia (binge/purge type), bulimia or no disorder. Images are not shared.

• Radiation Temporal Lobe Injury Diffusion MRI (SYSMH) (Patients' MRI scans, which are not shared)

  Diffusion MRI of patients with unilateral temporal lobe necrosis after radiotherapy for nasopharyngeal carcinoma, used to compare ADC, DTI, DKI, MAP-MRI and NODDI parameters between the injured and the healthy side. Only a table of per-lobe parameter values is public; the scans are not shared.

• TCD Schizophrenia Neuroimaging Genetics

  Structural, diffusion and resting-state brain MRI of patients with schizophrenia and healthy controls, from a Dublin study of genetics, cognition and emotion in schizophrenia. Images are not shared.

• Parent-Child Dyad Resting-State fMRI

  Resting-state brain fMRI and T1-weighted MRI of adolescents and their biological parents, collected to test whether parent-child similarity of intrinsic network connectivity tracks day-to-day mood synchrony. Raw images are not shared; group network maps are on NeuroVault.

• TICH-2 MRI Sub-study

  Brain MRI of patients with acute brain bleeding from a trial of tranexamic acid, used to test whether the drug causes small ischaemic lesions and later to develop microbleed detection. 219 patients, 115 with microbleed masks. Shared on written request.

• Tohoku normal Japanese brain database

  Brain MRI of healthy Japanese adults, collected to build age- and sex-specific reference brains and to study how gray and white matter change with age. 1,547 people aged 16 to 79 from Tohoku University. No public release or terms.

• TOP bipolar faces-matching fMRI (Oslo)

  Task fMRI with an emotional faces-matching paradigm plus T1w structural scans of 85 people with bipolar disorder and 121 healthy controls from the Oslo TOP study, analysed against diagnosis and polygenic risk. Raw images are available only on request; group z maps are on NeuroVault.

• TOPSY 7T psychosis cohort

  Ultra-high-field brain MRI and spectroscopy of people with untreated first-episode psychosis, established schizophrenia and controls, followed to relate glutamate, white matter and thought disorder to outcome. Longitudinal 7 T study in Ontario. Images are not public.

• TRAILBLAZER-ALZ 2 (donanemab phase 3 trial imaging)

  Safety brain MRI read for amyloid-related imaging abnormalities (ARIA), volumetric MRI and amyloid and tau PET from the donanemab phase 3 trial in early symptomatic Alzheimer's disease. 1,736 people. The images are not shared outside the sponsor.

• UCSF CMB_labeler SWI test set

  Unlabelled susceptibility-weighted brain MRI of patients with microbleeds caused by radiation therapy, shipped with a semi-automated tool for trying out microbleed detection. 10 volumes, 5 at 3 T and 5 at 7 T.

• UCSF Radiation-Induced Microbleed 7 T SWI Set

  Ultra-high-field susceptibility-weighted brain MRI of glioma patients with microbleeds caused by radiotherapy, used to train a network that separates true microbleeds from false detections. 91 scans of 73 patients at 7 T. No data release or sharing route.

• Ukrainian Refugees Trauma and Allostatic Load MRI Study (MRI scans, available from the authors on request)

  Structural brain MRI, questionnaires and stress biomarkers from a cross-sectional Wroclaw study of 60 Ukrainian refugees and 50 matched controls. Participant-level tables with MRI-derived measures are open under CC BY 4.0; the scans are available from the authors on request.

• Ultra-Low-Field MRI Brain Templates (Subject-level T1- and T2-weighted 64 mT MRI scans of the 100 participants.)

  Portable ultra-low-field brain MRI (T1- and T2-weighted, 64 mT) of healthy adults across the lifespan, averaged into age-group templates for spatial normalization. Only the templates, their segmentations and regional volumes are released; the scans of the 100 participants are not.

• Refractory epilepsy FDG PET/CT (Unicamp)

  Brain FDG PET/CT of surgical candidates with drug-refractory focal epilepsy at a university hospital in Campinas, Brazil, acquired under EEG monitoring to compare postictal with interictal localization of the epileptogenic zone. 110 patients. Images not shared.

• Ring-enhancing brain lesion MRI (UNESP/Unicamp)

  Brain MRI of patients with a ring-enhancing lesion from one Brazilian medical school hospital, used to tell inflammatory lesions from brain tumors with texture features and machine learning. 67 patients on one 1.5T scanner. Not shared.

• PPG (USC P01 neurovascular cohort)

  Brain MRI with dynamic contrast-enhanced imaging of blood-brain barrier leakage in older adults who are cognitively unimpaired or mildly impaired, collected to test whether vascular dysfunction drives early cognitive decline in APOE4 carriers. Held on LONI IDA without a public application.

• UW child trauma fear-conditioning fMRI

  Task fMRI and T1w brain MRI of Seattle children aged 8 to 16 with and without exposure to abuse or domestic violence, scanned during a fear-conditioning task to study aversive learning. 159 scanned, 147 analysed. Images not posted.

• VITA Vienna Transdanube Aging study

  Brain MRI, cognitive testing and blood markers of 75-year-olds born in 1925 and 1926 in Vienna, followed every 30 months to find which measures predict later dementia. Population-based birth cohort. Scans are not public.

• WRAP

  Cognition every two years plus brain MRI and amyloid and tau PET for subsets, from midlife adults mostly with a parent who had Alzheimer's disease, followed to study preclinical Alzheimer's disease. US cohort since 2001. Access by application.

• Yonsei/Gachon SWI Microbleed Dataset

  Susceptibility-weighted brain MRI with point labels of cerebral microbleeds, built to train and test a two-stage microbleed detector. 179 subjects from a Korean hospital with 760 microbleeds. Offered with the paper, but the authors withdrew the data in 2021.