Comparison of kernel functions in the prediction of cardiovascular disease in Artificial Neural Networks (ANN) and Support Vector Machines (SVM)

Cardiovascular diseases are currently the leading cause of death worldwide. There are challenges, such as untimely healthcare, lack of access to technologies and timely diagnoses. Therefore, this project focuses on the use of innovative tools, giving way to the need to use artificial intelligence in the field of Machine Learning to improve the prediction of cardiovascular diseases. The research focused on determining the most effective kernel function in Artificial Neural Network (ANN) and Support Vector Machine (SVM) algorithms, making a fair comparison and evaluating the accuracy and prediction time of each proposed kernel function. Based on the results, these new optimal kernel functions are integrated into the scikit-learn library, achieving validation in the appropriate configuration for predicting the risk of CVD. This innovative approach reduces detection time, minimising the chances of future complications from preventable diseases, and provides timely diagnosis and risk factors with early warnings that can be extremely useful for healthcare personnel.