Welcome to the new Gustave Roussy website
A new website designed to offer you a simpler, clearer, and more intuitive experience. Patients, caregivers, healthcare professionals, and donors: find information, news, and services more easily.
Biomathematics
- Thematic(s)
- Data & AI, Genetics / Genomics, Biomedical Imaging, Precision Medicine
- Attachment unit
- U1361 - Cancer Data Science
- Manager(s)
- Maria Vakalopoulo
- Institutional connection
-
Gustave Roussy, CentraleSupélec, Inserm, Paris-Saclay University
Summary
The “Biomathematics” research team, part of the “Cancer Data Science” unit, develops advanced mathematical, statistical, and computational tools to model, analyze, and interpret complex biological systems and clinical data. This team is dedicated to designing scalable and interpretable methods for extracting knowledge from heterogeneous, high-dimensional biomedical data. Its research covers mathematical modeling, statistical inference, machine learning, optimization, and artificial intelligence, with a particular emphasis on the integration of multimodal data such as genomics, spatial omics, medical imaging, digital pathology, and tabular data. The team tackles major challenges in cancer research, including modeling tumor heterogeneity, modeling disease progression, biomarker discovery, predicting treatment response, survival analysis, and patient stratification.
The methods developed by the team make it possible to transform complex cancer data into clinically actionable information, thereby helping to improve diagnosis, prognosis, and treatment selection. By integrating multimodal data and advanced modeling approaches, this research contributes to more accurate, personalized, and evidence-based clinical decision-making in oncology.
The team’s main areas of focus include the following:
-
Statistical and Mathematical Modeling: This research area focuses on developing statistical and mathematical models to characterize cancer biology and improve the interpretation of complex oncology data. By integrating multimodal data—including multi-omic data, imaging data, and clinical information—these approaches enable the identification of cancer-associated patterns, patient subgroups, and predictive biomarkers. Advanced mathematical modeling techniques facilitate the analysis of tumor heterogeneity, disease progression, and treatment response.
-
Data-Driven Methods for Precision Oncology: This research area focuses on developing data-driven approaches, leveraging machine learning and deep learning, to address the key challenges of precision oncology. These methods are designed to leverage large-scale, multimodal cancer data to improve prediction, classification, and decision-support tasks, including tumor characterization, patient stratification, biomarker discovery, and treatment response prediction.
Team members
-
Maria VAKALOPOULOU - Team Leader
Associate Professor, CentraleSupélec
ARAB Reda
Other status, including apprentice, CentraleSupélec
Hanna Bacave
Postdoctoral Researcher, CentraleSupélec
Malek Ben Salah
Other status, including apprentice, CentraleSupélec
BENKIRANE Hakim
Associate Professor, CentraleSupélec
BLAMPEY Quentin
Postdoctoral researcher, CentraleSupélec
BOUTAJ Sofiene
Other status, including apprentice, CentraleSupélec
Margherita BRUNO
Other status, including apprentice, CentraleSupélec
CHRAKI Imane
Other status, including apprentices, CentraleSupélec
CHRISTODOULIDIS Stergios
Associate Professor, CentraleSupélec
Paul-Henry COURNÈDE
Professor, CentraleSupélec
DAS Jyotishka
Other status, including apprentice, CentraleSupélec
Margaux DELIGNE
Other status, including apprentice, CentraleSupélec
EL YAALAOUI Adil
Other status, including apprentice, CentraleSupélec
FILLIOUX Léo
Other status, including apprentices, CentraleSupélec
FLORAKIS Konstantinos
Other status, including apprentices, CentraleSupélec
Vincent Fourmigue
Other status, including apprentices, CentraleSupélec
GELARD Maxence
Other status, including apprentices, CentraleSupélec
Irène Gentilini
Other status, including apprentices, CentraleSupélec
Félicie GIRAUD-SAUVEUR
Other status, including apprentices, CentraleSupélec
Gurvan HERMANGE
Associate Professor, CentraleSupélec
IAKOVLEVA Ekaterina
Other status, including apprentice, CentraleSupélec
KUGUSHEVA Alisa
Other status, including apprentice, CentraleSupélec
LETORT-LE CHEVALIER Véronique
Professor, CentraleSupélec
MALLEVAL Inès
Other status, including apprentices, CentraleSupélec
MAMANN Aaron
Other status, including apprentice, CentraleSupélec
Pierre MARZA
Postdoctoral researcher, CentraleSupélec
MAZET Paul
Other status, including apprentices, CentraleSupélec
MENARD Thomas
Other status, including apprentice, CentraleSupélec
Nabil Mouadden
Other status, including apprentices, CentraleSupélec
PERRAKIS Stelios
Other status, including apprentices, CentraleSupélec
PHAM Hugo
Other status, including apprentice, CentraleSupélec
RESTREPO David
Other status, including apprentices, CentraleSupélec
USUREAU Cédric
Other status, including apprentices, CentraleSupélec
Key publications
-
Blampey, Q., Benkirane, H., Bercovici, N., Mulder, K., Gessain, G., Ginhoux, F., André, F., & Cournède, P. H. (2025). Novae: a graph-based foundation model for spatial transcriptomics data. Nature Methods, 22(12), 2539–2550.
-
Moutakanni, T., Bojanowski, P., Chassagnon, G., Hudelot, C., Joulin, A., Lecun, Y., Muckley, M., Oquab, M., Revel, M.P., Vakalopoulou, M. “Advancing human-centric AI for robust X-ray analysis through holistic self-supervised learning.” Nat Commun (2026)
-
Giraud-Sauveur, F., Blampey, Q., Benkirane, H., Marinello, A., Cournède, P.-H., Christodoulidis, S. STHELAR, a multi-tissue dataset linking spatial transcriptomics and histology for cell type annotation. Sci Data 13, 665 (2026).
-
Blampey, Q., Mulder, K., Gardet, M., Christodoulidis, S., Dutertre, C.A., André, F., Ginhoux, F., and Cournède, P.H., 2024. Sopa: a technology-invariant pipeline for analyses of image-based spatial omics. Nature Communications, 15(1), p.4981.
-
Hermange G., Vainchenker, W., Plo I., Cournède, P.-H. Mathematical Modeling, Selection, and hierarchical inference to determine the minimal dose in IFNα therapy against Myeloproliferative Neoplasms. Mathematical Medicine and Biology. 2024; 41(2): 110–134.
Other scientific productions
Contacts
