Nonlinear Phenomena in Complex Systems, 2026, Vol.29, No.3, pp.292

NONLINEAR PHENOMENA IN COMPLEX SYSTEMS
An Interdisciplinary Journal

2026, Vol.29, No.3, pp.292 - 297

Comparative Analysis of Machine Learning Models for Predicting Cardiovascular Disease

Ramil M. Khusnutdinoff and Lilia M. Akhmetshina

This paper examines the application of machine learning methods to identifying cardiovascular diseases in patients and assessing the risk of heart attack. Four machine learning models were used to classify patients by health status: logistic regression, random forest, support vector machine, and artificial neural network. The selection of significant factors was carried out using the principal component analysis method. Optimal model architectures are proposed and hyperparameters are selected to ensure high classification accuracy. All models demonstrated high predictive power, as measured by precision, recall, F1 score, and AUC-ROC. The obtained results indicate the possibility of using the proposed approach as an effective tool for non-invasive diagnostics, patient classification and prognosis of cardiovascular diseases.

Key words: fully connected neural network, logistic regression, random forest, support vector machine, cardiovascular diseases

DOI: https://doi.org/10.5281/zenodo.23278251

Full text: Acrobat PDF (408 KB) Open Access