ARAMIS Lab
Building numerical models of brain diseases from multimodal patient data at Paris Brain Institute
The Aramis Lab brings together methodological researchers (computer scientists, applied mathematics) and medical experts (neurology, medical imaging) to build numerical models of brain diseases from multimodal patient data: medical imaging, clinical data and genomic data.
The Aramis Lab is a joint research team between CNRS, Inria, Inserm and Sorbonne University and belongs to the Paris Brain Institute (ICM), which is a neuroscience center based in the Pitié-Salpêtrière hospital in Paris, the largest adult hospital in Europe.
The team develops new data representations and statistical learning approaches that can integrate multiple types of data acquired in the living patient, including medical imaging, clinical and genomic data. In turn, these models shall allow for a better understanding of disease progression, and the development of new decision support systems for diagnosis, prognosis and design of clinical trials.
We are financially supported by these institutions.
Key methodological domains
Machine Learning & Representation Learning
Statistical learning, Deep learning, Generative models, Self-supervised learning, Bayesian models, Multimodal representation learning, Anomaly detection
Medical image processing
Magnetic Resonance Imaging (MRI), Positron Emission Tomography (PET), Image registration and synthesis, Clinical data warehouses
Longitudinal analysis
Non-linear mixed effects models, Riemannian geometry, Disease course mapping, Prediction-powered inference
Computational Pathology & High-Content Microscopy
Virtual staining, Explainable/Responsible AI, Physics-informed models, Digital histopathology, Spatial transcriptomics
Real-World Data, Reproducibility & Validation
Quality control pipelines, Benchmarking, Bias and reproducibility analysis, Administrative health databases (e.g. SNDS)
Main applications
Alzheimer's disease and fronto-temporal dementia
Clinical decision support systems, Disease progression models, Anomaly detection
Multiple sclerosis
Synthetic imaging biomarkers (e.g. PET-like biomarkers generated from clinical MRI)
Parkinson's disease, CADASIL, ataxias and ALS
Disease progression modelling, Clinical trial design, Care pathway analysis
Neuro-oncology
Automatic segmentation of brain tumours, Intraoperative applications and surgical planning
Neuro-ophthalmology
AI-based detection of retinal and neurological emergencies
2025 team retreat at la Ferme du Pignon, France