Research

Digital Organs

My research work is mainly focused on designing in silico models of human organs and medical image analysis methods using these models as prior knowledge and predictive tools. This knowledge can be at the anatomical, biomechanical or physiological levels.

To achieve this goal, I work in different scientific topics, like image processing, mesh generation, numerical analysis of physiological models, mesh adaptation to given data sets, parallel computing and graphical visualisation. To validate such models, I also work on the fusion of information from different modalities. As an in silico organ integrates anatomy, biomechanics and physiology, observations have to be made at these 3 levels as well.

Simulated isochrones visualised on the model of the heart,
same view as the measures from Durrer et al.

Image Processing

Medical imaging is a unique way to observe the human body in vivo. As the modelling of living tissues should be at the same scale as what is observable in vivo, it is really important to be able to integrate the clinical information from the medical images into the models.

The main work on medical imaging is extracting quantitative parameters or interest regions. My research on these tasks is mainly concentrated on segmentation with deformable models and preprocessing, like anisotropic filtering, for denoising.

Initial image, before anisotropic diffusion.

Final image, after 4D anisotropic diffusion.