Implementation of a Model for Risk Assessment of Cardiovascular Diseases Using Artificial Intelligence

In this work, we propose a model to help assess the risk of cardiovascular disease in an individual. The learning was carried out on a public dataset made available on the site Kaggle by Svetlana Ulianova and tested with a dataset of populations of West African countries. The first step was to extract a suitable data set for our study and pre-process it. Then a data analysis was done to determine the correlations between the attributes that make up our dataset. Finally, these data were explored with different algorithms based on decision trees such as the decision tree classifier, random forests to detect the most efficient. We have then studied the results and therefore concluded that the Decision Classifier Tree algorithm produced the best performance which is illustrated by the confusion matrix obtained, the performance-related metrics and a better prediction score.