Predicting the risk of cardiovascular diseases using multi-modal and multi-data types

In this internship, we will leverage the power of AI multi-modal algorithms to build a system for predicting the risk of cardiovascular (heart) diseases by extracting features from tabular data (clinical data and demographics) and the medical images in order to predict the risk of cardiovascular diseases at 5 years and 10 eyars. We have the access to a database of the Canadian Longitudinal Study on Aging (CLSA) database that consists of thousands of patients’ information (genomics, metabolomics, clinical, and imaging) with cardiovascular diseases. We will clean the data then we will use the patients’ clinical data combined with medical images to predict the risk of cardiovascular diseases. The multi modal algorithm will be built by deep learning technology using supervised learning. The target variable will be the occurrence of cardiovascular diseases at 5 years and 10 years. Eventually, we will compare our propositions with the published results (pool cohort equation and Framingham risk calculator), and interpret our innovation in manuscripts and presentations.

Faculty Supervisor:

Robert Avram

Student:

Partner:

Université Sidi Mohamed Ben Abdellah

Discipline:

Life Sciences

Sector:

Health and Related Sciences & Technology; Artificial Intelligence

University:

Université de Montréal

Program: