Research Projects

Computational pathology, biomedical imaging, multimodal learning and trustworthy AI for precision medicine.

Selected Projects

PhagoStat: Phagocytosis Unveiled

Mehdi Ounissi, Morwena Latouche and Daniel Racoceanu

A scalable and interpretable deep learning framework for neurodegenerative disease analysis, focusing on phagocytosis-related mechanisms and biomedical image interpretation.

Deep Learning Neurodegeneration Interpretability Biomedical Imaging
PhagoStat project overview

MALMO: Mathematical Approaches to Modelling Metabolic Plasticity and Heterogeneity in Melanoma

Janan Arslan, Ayse Gungor, Mehdi Ounissi, Pawan Kumar, Haocheng Luo, Matthieu Lacroix, Pierrick Dupré, Arran Hodgkinson, Christine Pignodel, Laurent Le Cam, Ovidiu Radulescu and Daniel Racoceanu

A collaborative project combining mathematical modelling, artificial intelligence and biomedical data analysis to better understand metabolic plasticity and heterogeneity in melanoma.

Melanoma Mathematical Modelling Metabolism AI for Cancer
WSI 3D reconstruction

Visual Deep Learning-Based Explanation for Neuritic Plaques Segmentation in Alzheimer’s Disease

Gabriel Jimenez, Anuradha Kar, Mehdi Ounissi, Léa Ingrassia, Susana Boluda, Benoît Delatour, Lev Stimmer and Daniel Racoceanu

A weakly supervised deep learning approach for whole-slide histopathological images, focusing on explainable segmentation of neuritic plaques in Alzheimer’s disease.

Alzheimer’s Disease Digital Pathology Weak Supervision Explainable AI
Attention U-Net architecture for neuritic plaque segmentation

STRATIFIAD: Refining Alzheimer’s Disease Patient Stratification

Gabriel Jimenez, Anuradha Kar, Mehdi Ounissi, Léa Ingrassia, Susana Boluda, Benoît Delatour, Lev Stimmer and Daniel Racoceanu

A computational histopathology project using explainable artificial intelligence to improve the stratification of Alzheimer’s disease patients from tissue-level features.

Computational Histopathology Patient Stratification Alzheimer’s Disease Precision Medicine
STRATIFIAD project illustration