Recomendação de um modelo de aprendizado de máquina parapredição de risco cardiovascular com biomarcadores da síndrome metabólica e Escore de Framingham

Abstract

The prediction of cardiovascular events in patients diagnosed with Metabolic Syndrome (MS) is a topic of great importance in the field of Health in general and crucial for Endocrinology. This dissertation aims to recommend a Machine Learning (ML) model to estimate the risks of cardiovascular events in patients with MS, exploring the markers of the Framingham Risk Score (FRS) and MS. Methodologically, we used a logistic regression model and conducted analyses with decision trees, random forest, gradient boosting, support vector machine, and k-nearest neighbors to test our central hypothesis that bioindicators (variables related to MS) exert a positive, strong, and significant impact on cardiovascular events in MS patients. Technically, the research was carried out through experiments conducted in different scenarios. In the first scenario, an algorithm was developed to assess cardiovascular risk in patients with and without MS. In scenarios 2, 3, and 4, patients with and without MS were analyzed, considering MS and FRSmarkersasdependentvariables,whiletheMetabolicSyndromeconditionwasadoptedasan independent variable. In the fifth scenario, an analysis was performed to select the most suitable regression and classification model for predicting cardiovascular risk in a combined dataset of heart diseases. In the sixth scenario, the developed model was based on the Framingham Global Risk Score (FRS), incorporating MS markers into the experiments. Data were obtained from the National Center for Health Statistics (NHANES) repository, a combined dataset of heart diseases from the UCI Machine Learning Repository, and the Kaggle platform. In summary, the main findings of this dissertation are as follows: In the first scenario, a percentage difference of 81.74% was observed in mean CVD Risk between populations with and without Metabolic Syndrome, demonstrating a significant increase in cardiovascular risk in the population with MS. In scenarios two, three, and four, the Random Forest (RF) model excelled, achieving high accuracy in all variable combinations, especially in the combination of MS markers with gender. In the fifth scenario, the RF model was identified as the most suitable, emphasizing the importance of variables related to MS in predicting cardiovascular risk and highlighting the need for model improvements to better identify positive cases. Both the model with four MS markers and the model with five MS markers combined with the Framingham Risk Score (SM + FRS) demonstrated considerable performance, with significant correlations and accuracies (0.80 and 0.84, respectively). These simpler variable combinations can be an interesting approach as they provide relevant information for predicting cardiovascular risk in a less invasive manner, avoiding the need for more complex tests.