My
research work concerns the optimal exploitation of the content of
multimodal medical images (PET/CT, MRI, CT...) in the context of the
treatment of patients with cancer, for diagnostic and prognostic
applications, therapeutic planning and evaluation.
From a methodological point of view, the tools I use relate to image
processing and analysis and statistical learning for designing of
multiparametric models. The aspects on which I have particularly
focused in recent years are new neural network architectures and
interpretability techniques for so-called "black box" models such as
deep neural networks.
