Grant: Project Grant

Parkinson's disease (PD) is a severe neurodegenerative disease that affects more than 2% of the Canadian population above 65, contributing to societal and financial burden. Despite a large body of research in the literature, prediction of the disease progression is still an unsolved problem. The lack of efficient biomarkers for PD progression is likely due to several factors 1) the difficulty of assembling large datasets of controls and PD patients with multiple time points that include all the relevant brain imaging, genetic and clinical or cognitive assessments, 2) the lack of multimodal predictors that account for the multiple features of the pathology, including precise brain imaging assessments, and 3) the lack of validation and assessment of the PD progression prediction on an unseen population. We propose to develop accurate predictors of individual PD progression using machine learning techniques that include information such as brain imaging (in particular structural and diffusion imaging), cognitive and clinical assessments, and genetic information. We will assemble a unique set of harmonized data on a diverse population from international and local sources to help address the PD progression prediction at the individual level. We will develop progression biomarkers validation techniques to ensure the robustness of PD progression prediction and offer clinicians tools for accurate view of a patient's disease evolution. Accurate prediction of PD progression will assist in the selection of patients for clinical trials, and help develop more targeted therapeutic strategies.