https://doi.org/10.1140/epjs/s11734-024-01451-5
Review
Prediction of cardiovascular events with using proportional risk models and machine learning algorithms: a systematic review
1
Federal State Budgetary Institution National Medical Research Center for Therapy and Preventive Medicine of the Ministry of Healthcare of the Russian Federation, Petroverigsky Per., 10, Building 3, 101990, Moscow, Russia
2
State Healthcare Institution of the Tula Region “Kireevskaya Central District Hospital”, 44 Lenin St, 301260, Kireevsk, Russia
3
Therapy and General Medical Practice of the Russian Federation Health Ministry, Moscow, Russia
4
Sciences Russian Academy, Moscow, Russia
Received:
29
October
2024
Accepted:
12
December
2024
Published online:
7
May
2025
Each year, ever more people fall victim to cardio-vesicular disorders. Today, the main tools for predicting cardio-vesicular risks are scores based on proportional risk models (Cox regressions). However, over the last years, many researchers have come to the conclusion that the use of machine learning and artificial intelligence will help to improve the quality of predicting an occurrence of adverse cardio-vesicular conditions. A range of literature, including 58 research papers using cardio-vesicular risk evaluation techniques based on Cox regression and machine learning, has been systematically reviewed. Prediction capabilities of machine learning are superior to traditional linear methods of data analysis. The mean values of AUC (area under the curve) are, respectively, 0.82 and 0.75, while p = 0.003. Also, it became possible to single out the most often used and effective prediction algorithms. Those appeared random forest, gradient boosting and deep learning. However, unlike the traditional prediction scores, 80% of the presented machine learning algorithms did not pass the external validation on independent samples. In addition, the use of machine learning calls for a large amount of quality digital data. Machine learning is a prospective method of predicting cardio-vesicular conditions, whose treatment calls for switching to electronic accounting of medical documentation and aggregation of a large amount of high-quality structured information.
Copyright comment Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
© The Author(s), under exclusive licence to EDP Sciences, Springer-Verlag GmbH Germany, part of Springer Nature 2025
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
