Utilisation de l'intelligence artificielle pour identifier les marqueurs cognitifs et cérébraux de la démence dans la maladie de Parkinson et son stade prodromal


Loubna Mekki BERRADA

Université de Montréal


Domaine : vieillissement

Programme Formation de maîtrise

Concours 2018-2019

Partenaire :

Parkinson Québec

Dementia with Lewy bodies (DLB) and Parkinson's disease (PD) are characterised by major motor deficits, psychiatric symptoms and cognitive deficits that limit the autonomy and impact the lives of the people affected and their relatives. Rapid eye movement (REM) sleep behavior disorder (RBD) is now well known as an important risk factor for these two neurodegenerative diseases. It's a sleep disorder characterised by abnormal and undesirable behaviors during the REM sleep. Therefore, longitudinal studies of RBD patients offer an interesting way to determine the predictive factors of the development of PD or DLB. This represents an ideal challenge for a prediction approach based on machine-learning. However, to date, artificial intelligence has never been used on longitudinal data to identify RBD patients and their progression towards DLB or PD.

The objective of this project is to use artificial intelligence to determine the contribution of cognitive data and EEG recordings taken at baseline (T0), in distinguishing healthy controls from patients, and predicting among them those who will progress (T1) towards dementia or PD. The use of supervised learning will allow us to identify the best tests to administer and the most relevant neuronal markers, through their predictive performance. Thus, this research project will result in, not only an advancement of knowledge, but also to prediction tools that will make possible to detect at an early stage RBD patients at high risks of evolving towards a neurodegenerative disease. Lastly, in the medium term, it will enable the objective study of the effectiveness of neuroprotective treatments aiming at slowing or preventing the development of neurodegenerative diseases.