WHO WE ARE
volBrain is an academic, free-access cloud platform for medical image analysis created by the Image Processing and AI group at ITACA (Universitat Politècnica de València), with CNRS / University of Bordeaux (LaBRI). Since 2015 it has made research-grade MRI analysis available to clinicians and researchers who lack computing resources or technical expertise: upload an image, get a standardized quantitative report (PDF, CSV and NIfTI segmentations) in about 20 minutes. The volbrain.net platform has processed more than 800,000 cases for over 1,600 users in more than 4,600 institutions worldwide, and more than 800 publications have used our tools. It is funded by Spanish (MINECO/MICIU) and French (ANR: DeepvolBrain, TRAIL) national programmes.
WHAT WE OFFER (Partner seeks Consortium/Coordinator)
We offer expertise as a Partner or work package leader in medical image analysis and AI-based imaging biomarkers, with ready-to-use, validated pipelines:
-
vol2Brain / volBrain: whole-brain parcellation (>130 structures), tissues, cortical thickness, white matter lesions.
-
AssemblyNet (+ AD and AD-FTD versions): deep learning parcellation (> 130 strctures) and prediction/differential diagnosis of Alzheimer’s disease and frontotemporal dementia.
-
lesionBrain / DeepLesionBrain: white matter lesion segmentation on T1w and FLAIR, with disconnectome maps (e.g. multiple sclerosis).
-
cognitiveTree and motorTree: decision support for differential diagnosis of cognitive and motor disorders.
-
petBrain: amyloid, tau and neurodegeneration status from MRI and PET.
-
HIPS, DeepThalamus, DeepCERES, pBrain: hippocampal subfields, thalamic nuclei, cerebellar lobules, deep and subthalamic nuclei.
-
BrainStructureAges: brain age and age per structure as a biomarker of ageing and disease.
VALUE PROPOSITION AND COMPETITIVE ADVANTAGE
Our pipelines turn raw MRI into reproducible volumetric and diagnostic biomarkers without local infrastructure, which lowers the barrier for hospitals and trials. Compared with typical academic tools, they are already deployed as a service, validated in peer-reviewed publications, robust to multi-centre and multi-scanner data, and fast. Our large user community provides continuous real-world feedback and visibility for any project we join.
MATURITY
Our core pipelines (volBrain, vol2Brain, AssemblyNet, lesionBrain) are mature and operate as a public service used worldwide. Newer diagnostic tools (cognitiveTree, motorTree, petBrain, AssemblyNet-AD/FTD) are validated on research cohorts and ready for prospective clinical validation.
WHAT WE CAN CONTRIBUTE TO A CLUSTER 1 HEALTH PROPOSAL
-
Leadership of an imaging / AI analysis work package.
-
Adaptation and validation of our pipelines on project cohorts (neurological, neurodegenerative, neuropsychiatric, neuro-oncological), and extension to new modalities and organs.
-
Imaging biomarkers for early diagnosis, patient stratification, progression and treatment-response monitoring in clinical studies and trials.
-
Scalable cloud processing with harmonized outputs ready for statistical analysis, privacy-preserving deployment in collaboration with clinical partners.
-
Methodological research: domain adaptation and harmonization across sites, uncertainty estimation, explainable AI, normative and brain-age models, learning with limited annotations.
TEAM
Led by Prof. José V. Manjón (UPV) and Dr. Pierrick Coupé (CNRS, Univ. Bordeaux), with a long-standing record in image processing, deep learning and neuroimaging, and close ties with clinical groups.
WHO WE ARE LOOKING FOR
-
Coordinators building 2027 Cluster 1 Health consortia that need an imaging analysis partner.
-
Hospitals and clinical research centres with imaging cohorts who want to validate biomarkers in real clinical settings.
-
Pharma, biotech and medtech companies (including SMEs) running neuroscience trials or seeking to integrate imaging biomarkers in products.
-
AI, radiomics, genomics and digital health groups for multimodal approaches.
Contact: team@volbrain.net | www.volbrain.net