Machine learning based on laboratory data for disease prediction

Objective : to review domestic and foreign literature on the issue of machine learning methods applied in medical information systems (MIS), analyze accuracy efficiency technologies under study, their advantages disadvantages, possibilities implementation clinical practice. Material . The search was performed PubMed/MEDLINE databases covering period from 2000 2020 (using groups keyphrases: "machine learning", "laboratory data", "clinical events", "prediction diseases"), CyberLeninka ("machine diseases" Russian keyphrases combinations) Papers With Code ("clinical diseases", "electronic health record"). After reviewing full text 30 sources that met selection criteria, 19 most relevant articles were selected. Results An analysis describe application artificial intelligence techniques obtain predictive analytics, taking into account about patients, such as demographic, anamnestic, laboratory data, data instrumental studies, existing former diseases available MIS, performed. ways predicting adverse outcomes using considered. Information significance used for constructing high-precision mathematical models is presented. Conclusion Implementation algorithms MIS seems be a promising tool effective prediction events wide real It corresponds global trend development personalized medicine based calculation individual risk. There an increase activity research field noncommunicable technologies.