BRAINTWIN
Brain Research through Advanced INTegration with Wide-scale Intelligent Networks
Building digital twins for precision neurology through multimodal AI, federated learning, and advanced mathematical modeling for brain tumors and cerebral small vessel disease.
Digital Twins for Precision Neurology
Revolutionizing brain disease treatment through AI-driven personalized medicine
BRAINTWIN is a four-year research project (2025-2029) funded by the French National Research Agency (ANR) as part of the PEPR Sante Numerique under France 2030. The project aims to create patient-specific digital twins for two major neurological conditions:
-
1
Primary Brain Tumors
Gliomas, meningiomas, and primary CNS lymphomas
-
2
Cerebral Small Vessel Disease (SVD)
Leading cause of stroke and contributor to 45% of dementia cases
Multimodal AI
Integrating MRI, histopathology, omics, and clinical data
Federated Learning
Privacy-preserving AI across hospitals
Foundation Models
Transfer learning from large biomedical datasets
Clinical Validation
Real-world deployment and evaluation
Work Packages
Our research is organized into seven interconnected work packages
Data Integration & FAIR Platform
Building a secure, FAIR-compliant platform for aggregating and harmonizing multimodal brain tumor and SVD datasets from multiple institutions.
Multimodal AI Methods
Developing novel AI algorithms to extract features from each modality and fuse them into unified patient representations using foundation models.
Federated Learning & Privacy
Adapting federated learning methodologies to multimodal data, enabling privacy-preserving training across multiple hospital sites.
Clinical Validation: Neuro-Oncology
Rigorous evaluation of the digital twin in brain tumor patients through retrospective and prospective studies at partner hospitals.
Clinical Validation: SVD
Testing and validating the digital twin approach for cerebral small vessel diseases and related neurodegenerative disorders.
Dissemination & Management
Maximizing project impact through open science, stakeholder engagement, and ensuring smooth project governance and coordination.
Key Innovations
Four major innovative features drive the project
Explainable Multimodal Fusion
Deep learning models integrating 3D MRI, whole-slide histology, clinical text, and genomic markers with cross-attention transformers and saliency mapping for interpretable predictions.
Resilient Federated Learning
GDPR-compliant training across decentralized hospital networks using robust aggregation algorithms (FedProx, Krum) and differential privacy techniques.
Innovative Generative Modeling approaches
Advanced generative models for modeling longitudinal disease courses, imputing missing modalities, and synthesizing biologically realistic data to enhance training.
Cross-Domain Clinical Validation
Rigorous validation in both neuro-oncology and cerebrovascular disease using retrospective and prospective studies with real-world clinical outcomes.
Consortium Partners
A multidisciplinary collaboration across leading French institutions
Latest News
Stay updated with BRAINTWIN developments
Project Kickoff
BRAINTWIN officially launches with the kickoff meeting bringing together all consortium partners in Paris.
Read MoreANR Funding Approved
BRAINTWIN receives funding approval from ANR under the PEPR Sante Numerique program of France 2030.
Read MorePhD Positions
Multiple PhD positions will be available across partner institutions. Check back for updates.
Contact UsInterested in Collaborating?
We welcome collaborations with researchers, clinicians, and industry partners working on digital health and AI for neurology.