PhD in AI and Syndromic Surveillance / Time Series Foundation Models at Santé publique France Santé publique France in France

PhD opportunity in AI, syndromic surveillance, time series foundation models, anomaly detection, forecasting, epidemiology, public health, data science, and statistics at Santé publique France, with academic co-supervision involving Matthieu HANF, Camille Pelat, and Anne Fouillet.

The thesis focuses on evaluating and improving foundation models for time series (TSFMs) to support early detection and forecasting of health signals in syndromic surveillance. The work uses both simulated datasets and real surveillance data from the SurSaUD / OSCOUR system, with applications to large-scale automated health signal monitoring.

Research themes include comparison of TSFMs with classical statistical, machine learning, and deep learning approaches; anomaly detection; epidemiological forecasting; and optimization using covariates, pretraining, and advanced preprocessing/postprocessing. The project is hosted in the DATA department of Santé publique France, with one day per week in the IPLESP team on surveillance and modeling of transmissible diseases.

Eligibility: candidates should hold an M2 or engineering degree with a research internship in data science/statistics, have strong programming skills in Python and R, and demonstrate a strong interest in public health. Good intermediate-upper level French and English are required.

Funding details are not specified in the post. The application deadline is 8 May 2026, and the PhD is expected to start in autumn 2026.