Artificial intelligence in the prevention and diagnosis of cardiovascular diseases in Russia (literature review)

Aim: To assess work carried out in Russia over the past 5 years to identify the risks of developing cardiovascular diseases using artificial intelligence (AI) methods and technologies. Materials and methods: A systematic review of the available literature over the past 5 years on the use of machine learning and knowledge representation methods in predicting the development and outcomes of cardiovascular diseases in Russia was carried out based on the Prisma methodology. 221 articles were analyzed. Results and discussion: The result of the systematic review is an analysis of the presented methods of model building, which ones are most often used, and with the help of which metrics researchers evaluate the quality of the obtained model. Machine learning methods are used most frequently compared to knowledge-based methods (expert systems), 22 articles and 7 articles respectively. Analysing the machine learning methods used, it can be noted that the first 5 places among the methods used in Russia are occupied by neural networks, regression, decision tree, boosting and random forest. Among the models of knowledge representation, ontology and semantic networks, which are often used for structuring and analyzing complex data in various knowledge domains, turned out to be the most widespread in the presented works. Almost all researchers in their papers evaluated the created model on a test sample and considered numerical metrics: accuracy (accuracy of measurement), precision (accuracy of the measuring instrument), completeness (recall), specificity (specificity), sensitivity (sensitivity), AUC (area under the ROC curve), F-measure (F-measure). The discussion is a discourse on the use of different metrics to evaluate different model variants. Conclusion: The results of the analysis of works using AI for the prevention and diagnosis of cardio-vascular diseases are summarized, and an assessment of their further application is given.