Mélodie Monod

I earned a BSc in Statistics from the University of Geneva and an MSc in Statistics with Distinction from Imperial College London. I completed my PhD in Modern Statistics and Statistical Machine Learning at Imperial in 2023, focusing on Bayesian models and methods. I then joined Novartis as a Principal Biostatistician, leading the development of advanced statistical methods, including deep learning for survival prediction. Since October 2024, I have been a postdoctoral researcher, continuing work at the intersection of Bayesian statistics, Statistical Machine Learning, and real-world applications. 

My research focuses on advancing deep probabilistic generative modeling with an emphasis on statistical foundations and real-world predictive inference. I work with conditional denoising diffusion models, stochastic interpolants, and flow/bridge matching techniques to develop flexible and scalable generative frameworks. These models enable the synthesis of complex high-dimensional data distributions and support downstream tasks such as counterfactual reasoning and structured prediction. A key application of this work lies in medical imaging for the Oxford–Novartis Collaboration for AI in Medicine. 

I am based at the Big Data Institute and the Department of Statistics.