Cardiovascular disease prediction by machine learning algorithms based on cytokines in Kazakhs of China
Clinical Epidemiology Jun 12, 2021
Jiang Y, Zhang X, Ma R, et al. - The present study was conducted to systematically explore the utility and performance of 7 widely used machine learning (ML) algorithms in predicting cardiovascular disease (CVD) risks. Researchers enrolled 1508 Kazakh individuals in China without CVD at baseline who completed follow-up. Individuals were assigned randomly into the training set (80%) and the test set (20%). They used L1-penalized logistic regression, support vector machine with radial basis function, decision tree, random forest, k-nearest neighbors, Gaussian naive Bayes, and extreme gradient boosting for prediction CVD outcomes. During model development and hyperparameters tuning in the training set, ten-fold cross-validation was used. The results demonstrate that in the Kazakh Chinese population, support vector machines with radial basis function and logistic regression can be applied to guide clinical decision-making, and there is a need for future study to ensure their accuracy.
-
Exclusive Write-ups & Webinars by KOLs
-
Daily Quiz by specialty
-
Paid Market Research Surveys
-
Case discussions, News & Journals' summaries