Research Profile- Dr. Simon Duchesne

Print

Research Summary for Dr. Simon Duchesne

 

Affiliation: 

Quebec City Mental Health Institute (researcher), Radiology Dept., Faculty of Medicine, Laval University (Assistant Professor,  Research Co-Director), Quebec neurosciences and mental health imaging consortium (executive director) 

Contact: 

Tel.: (418) 663-5741 poste 4777, Fax: (418) 663-5971, email:  This email address is being protected from spambots. You need JavaScript enabled to view it.  

 

Description of Research:

According to some estimates there are more than 500,000 people with dementia in Canada, or one elderly out of eleven. Alzheimer's disease is responsible for more than two-thirds of these cases. This number is expected to grow in the coming decades as the population ages, bringing with it significant economic and societal consequences for these individuals, their caregivers and the community at large. 

The overlap in cognitive symptoms between normal aging and those associated with Alzheimer's disease, especially in its early phase, complicates its clinical diagnosis. Clinicians need additional tools and methodologies to measure this decline quantitatively and thus achieve a clinical diagnosis not only with higher confidence, but as early as possible, before patients have crossed the threshold of clinical dementia.

Dr. Simon Duchesne, eng., Ph.D., founded the MEDICS Laboratory at the Centre de recherche de l’Institut universitaire de santé mentale de Québec, to address this problem by proposing new techniques for analysis of brain magnetic resonance images. This noninvasive neuroimaging modality provides anatomic images of the human brain in vivo with high resolution. MEDICS researchers were among the first to use machine-learning algorithms to extract image features that are specific to Alzheimer's disease, and can therefore serve as putative biomarkers. Lately, Dr. Duchesne and his team have taken to study dynamic models of biomarkers, beyond imaging, to arrive at a better description of the pathological process related to Alzheimer’s.

To accomplish this work, the MEDICS Laboratory can count on a powerful ally: CLUMEQ’s supercomputer Colossus at Laval University. Their image processing system is located on Colossus, which allows them to make their calculations on databases of hundreds or even thousands of images simultaneously. For example, the process of identifying the hippocampus on 800 magnetic resonance images, an important structure of the human brain involved in the early stages of Alzheimer's disease, takes over 140 hours of computation on computers in the lab, but could theoretically be completed on Colossus in less than a half day. Most recently, Dr. Duchesne and Colossus have agreed on the deployment of a GPU architecture to further accelerate some of the tasks assigned.

More importantly perhaps, Colossus and CLUMEQ allows MEDICS researchers to dream the next generation of highly parallelized algorithms, which may disrupt existing paradigms in medical image processing and allow the computation of results inaccessible to date.

 

Publication:

http://iopscience.iop.org/1742-6596/341/1/012011