Symposium scientifique CRESS « How AI and LLM are transforming research in Epidemiology? »

Symposium scientifique CRESS « How AI and LLM are transforming research in Epidemiology? »

Cet événement a pour objectif d’explorer les transformations majeures que l’intelligence artificielle (IA) et les grands modèles de langage (LLM) opèrent dans le domaine de la recherche épidémiologique : de la construction des preuves à la synthèse des connaissances, en passant par l’évaluation par les pairs.

Experts nationaux et internationaux / Exposition posters doctorants CRESS

  • Localisation : Site Odéon Université Paris Cité
  • Date : mardi 09 décembre 2025
  • Inscription obligatoire : ici

Programme

  • 9:45-10:30 : Keynote Speech. Can AI and LLM generate trustworthy knowledge? The example of mathematics. Ivan Nourdin (Professor in Mathematics, University of Luxembourg).

Ivan Nourdin is full Professor in Mathematics at the University of Luxembourg. He is also the Head of the research group Exploring the Fascinating World of Unpredictable. His research interests include Malliavin calculus, Stein’s method, functional inequalities, free probability, rough paths theory, and inference for high-dimensional problems. Ivan Nourdin is author/coauthor of more than 110 publications in international scientific journals as well as 2 monographs. He joined the Université Pierre et Marie Curie (Paris VI) as an assistant professor in 2005 and then the Université de Lorraine as a full professor in 2010. In 2011, he was awarded the Annual Prize of the Fondation des Sciences Mathématiques de Paris. In 2013, he was awarded the France Scopus Researcher Award in the field of Mathematics by Elsevier. He was appointed professor of stochastic modelling at the University of Luxembourg in March 2014. Since 2018, he has served as Director of the Bachelor’s programme in Mathematics. In 2015, together with Giovanni Peccati he was awarded the FNR Award for Outstanding Scientific Publication (for the book “Normal Approximations with Malliavin Calculus: from Stein’s method to universality”, published by Cambridge University Press in 2012).

  • 10:30 – 10:45 : Opportunistic screening for osteoporosis. Christian Roux (Professor of Rheumatology, Université Paris Cité)
  • 10:50 – 11:10 : Artificial intelligence–assisted otoscopy. Jérémie Cohen (Professor of Paediatrics, Université Paris Cité)
  • 11:10 – 11:30 : AI for participatory research in nutrition: ongoing solutions and future perspectives. Alice Bellicha (Associate Professor of Nutrition, Université Sorbonne Paris Nord)
  • 11:30 – 11:50 : Use of machine learning in accelerometry research. Ian Danilevicz (Postdoctoral researcher, Inserm-Université Paris Cité)
  • 11:50 – 12:10 : What does pain mean? When digital and medical humanities meet to highlight a public health challenge. Astrid Chevance (Professor of Epidemiology, Université Paris Cité)
  • 12:10 – 12:30 : Synthetic data for augmented trials. Alex Fernandes (PhD Student, Université Paris Cité)
  • 14:00 – 15:00 : Keynote Speech. Kamran Abbasi (Editor in chief of The BMJ)

Kamran Abbasi is the Editor-in-Chief of the BMJ. He is a physician, journalist, editor, and broadcaster. After spending five years in hospital medicine across various specialties such as psychiatry and cardiology, he worked at The BMJ from 1997 to 2005. He has now returned to The BMJ as Editor-in-Chief for content, responsible for leading the journal’s strategic growth internationally, both digitally and in print. Throughout his career in medical editing, Kamran has served as Acting Editor-in-Chief and Deputy Editor of the BMJ, Editor-in-Chief of the Bulletin of the World Health Organization, and Consulting Editor for PLOS Medicine. He is also Editor-in-Chief of the Journal of the Royal Society of Medicine and JRSM Open. He created three major online learning resources for doctors’ professional development, including BMJ Learning and the Royal Society of Medicine’s video conferencing service. Kamran has held board-level positions and served as Chief Executive of an online learning company. He has been a consultant for several major organizations, including Harvard University, the UK’s NHS, the World Health Organization, and McKinsey & Co. In addition, Kamran is an honorary visiting professor in the Department of Primary Care and Public Health at Imperial College London. He is a Fellow of the Royal College of Physicians of Edinburgh and the Royal College of Physicians of London, a patron of the South Asian Health Foundation, and a member of the King’s Fund General Advisory Council.

  • 15:00 – 15:45 : Joerg Meerpohl (Professor and Director of the Institute for Evidence in Medicine at the Medical Center & Faculty of Medicine, University of Freiburg)

Professor Joerg Meerpohl is Director of the Institute for Evidence in Medicine at the Medical Center and Faculty of Medicine of the University of Freiburg, Germany, and founding Director of the Freiburg GRADE Center, established in 2013. He is trained as a pediatrician, hematologist, and pediatric oncologist. For more than ten years, Professor Meerpohl has been an active member of the GRADE Working Group and has served as a GRADE methodology advisor for several WHO expert groups and the Robert Koch Institute in Germany, among others. Professor Meerpohl is also Director of Cochrane Germany. He is currently an elected member of Cochrane’s Governing Board and of the GRADE Guidance Group. His main research interests include systematic review methodology, research transparency, and guideline methodology. Professor Meerpohl has published over 350 articles indexed in PubMed.

  • 15:45 – 16:15 : Three proofs of concept for accelerating and simplifying research tasks using AI. Viet Thi Tran (Professor of Epidemiology, Université Paris Cité)

Artificial intelligence (AI) is increasingly explored as a means to accelerate and transform research workflows. In this presentation, we introduce three proof-of-concept applications of AI designed to reduce time, cost, and operational burden across the research lifecycle. First, we will describe our work on systematic reviews, where AI was initially used to accelerate screening but has since enabled a more radical rethinking of the process by generating evidence syntheses without a traditional search strategy. Second, we will share experimental findings on the use of AI to streamline the analysis of open-text data. Third, we will highlight how we used AI to significantly reduce the low-value technical and administrative workload that precedes study initiation.

Pr Viet-Thi Tran is a Full Professor of Epidemiology at Université Paris Cité and at the Assistance Publique Hôpitaux de Paris (AP-HP), in Paris France. He leads a research group around novel methods to generate primary evidence in the METHODS team, which is specialized on therapeutic evaluation in chronic conditions. His work revolves around three topicss: 1) minimally disruptive medicine (i.e., how to care for patient in a way that respects his life goals), and especially how digital tools may help achieve such minimally disruptive medicine; 2) citizen science methods (i.e., how to involve patients in research and in evidence generation); and 3) novel methods in therapeutic evaluation (e.g., target trial emulation, virtual controls, etc.)

Isabelle Boutron (Professor of Epidemiology Université Paris Cité, Director of the CRESS)

Back to top