The "Prospects for the using of machine learning methods for predicting cardiovascular diseases" article by Gusev A. V., Gavrilov D. V., Korsakov I. N., Serova L. M., Novitsky R. E., Kuznetsova T. Yu. was published in the 3rd issue of the magazine "Information technologies for the Physician 2019".
Morbidity and mortality from cardiovascular diseases (CVD) have remained the leading rate in recent decades worldwide. Primary prevention methods based on the management of cardiovascular risk factors are the most effective in reducing the burden of CVD. In preventive medicine, the instruments for risk measurement are used for risk management of CVD: scores that were obtained as a result of long prospective studies. However, the practical application of the developing scales has shown limitations in the forecast accuracy.
Machine learning makes it possible to improve the accuracy of cardiovascular risks prediction due to nonlinear relationships of their deeper adjustment between risk factors and disease outcomes.
2236 patients’ data were used. We trained the model on the features used in the Framingham score construction. We compared the resulting model and the Framingham score for the accuracy of the cardiovascular event prediction. Thus, according to the ROC analysis for the Framingham score, the indicators are as follows: precision Accuracy: 70,0%, the AUC: 0.59. At the same time, for the machine learning model, similar indicators were: Accuracy: 78,8%, AUC: 0.84.
Thus, the use of machine learning algorithms, including deep learning algorithms, can significantly improve the accuracy of cardiovascular risk prediction of trained models.