Nikhil Potluri
Vol. 17, Jan-Jun 2024
Abstract:
Cardiovascular diseases (CVDs) are a major cause of death worldwide, ranking among the deadliest disease. By utilizing statistical and machine learning (ML) algorithms to discover risk biomarkers, CVDs can be early detected and prevented. In this work, we use biochemical data and clinical CVD risk factors to predict CVD-related death within a 10-year follow-up period using machine learning models like Logistic Regression (LR), Support Vector Machine (SVM), Random Forest (RF), Naïve Bayes (NB), Extreme Grading Boosting (XGB), and Adaptive Boosting (AdaBoost). Using the Ludwigshafen Risk and Cardiovascular Health (LURIC) study cohort, we included 2943 individuals in our analysis, of whom 484 were declared deceased from cardiovascular disease. For every model, we determined its accuracy (ACC), precision, recall, F1-score, specificity (SPE), and area under the receiver operating characteristic curve (AUC). According to the comparative analysis's results, the most dependable algorithm is logistic regression, which has an accuracy of 72.20%. In the TIMELY trial, these findings will be utilized to calculate the risk score and mortality of cardiovascular disease in patients with a 10-year risk.
DOI: http://doi.org/10.37648/ijrmst.v17i01.010