Faculté des sciences et de génie
Philippe Després is a full professor in the Department of Physics, Engineering Physics and Optics and member of the Cancer Research Center of Université Laval, medical physicist at CHU de Québec and a regular researcher at its affiliated Research Center, membrer of the Institute intelligence and data of Université Laval and member of the Researcher Council of the New Digital Research Infrastructure Organization (NDRIO). He is the designated principal investigator of the RHHDS program.
He was trained at Université Laval (MSc 2000, Physics), Université de Montréal (PhD 2005, Physics) and University of California, San Francisco (postdoc 2005-2007, Biomedical Engineering, Molecular Imaging).
Professor Philippe Després is involved in several projects encompassing hardware and software aspects of medical imaging modalities, notably low-dose X-ray imaging, advanced imaging techniques, and solid-state detectors for molecular imaging. He pioneered high-performance computing (HPC) approaches with commodity graphics hardware (GPUs) that led to innovative applications in image processing/reconstruction and radiation dose calculations, including a fast GPU-based Monte Carlo engine to simulate energy transport in matter (GPUMCD).
As a HPC expert, professor Philippe Després is also involved in data-driven research approaches, data infrastructures and FAIR-compliant research data management. In this regard, he is responsible of biomedical data at CHU de Québec of Université Laval Research Center, the data architect of the PULSAR health research platform, the assistant director of the Big Data Research Center at Université Laval, and the co-lead of the sustainable health axis of the Observatoire international sur les impacts sociétaux de l’intelligence artificielle et du numérique (OBVIA).
Supervised classification allows to build predictive models based on complex data to help human decision making processes. It has undergone an impressive development in recent years, particularly thanks to neural networks and the use of big data. However, these methods are not relevant to use on databases in which only a few instances are available to build the model, and even less when these instances are described by a large number of features.
In medical imaging, radiomic features make it possible to characterize heterogeneity of a region of interest at the anatomical level. This way of quantifying the heterogeneity of a region of interest can be useful, for example, in order to identify the more aggressive tumors in oncology. To do this, we hypothesize here that variation in magnetic resonance imaging (MRI) acquisition sequences and its resulting different levels of contrast would make it possible to optimize the subsequent radiomic analysis.
This project consists of establishing good practices in health data management and building a software infrastructure in order to apply them.
We have developed pipelines that allow daily recovery of brachytherapy treatment data in order to calculate and store their dosimetric indices in a database dedicated to research. These indices are essential for planning radiotherapy treatments and for estimating their quality.