OBJECTIVE
To develop prognostic models of the individual risk of developing chronic non-communicable diseases (CNCDs) with integration of sociomedical determinants in the population of persons older than working age (POWA) using artificial intelligence.
MATERIAL AND METHODS
Modeling and prediction of CNCDs in POWA through the use of artificial intelligence consisted of studying the prevalence of CNCDs per surveyed person by performing preventive medical examinations of 5170 people aged 60—95 years, analyzing the prevalence of risk factors in POWA with CNCDs, formation of the variance complexes, quantification of the effect size (η²) of risk factors on the development of CNCDs. In addition, calculations of normalized intensity indicators, relative risk and prognostic coefficient were carried out. Individual prediction matrix and software for artificial intelligence were developed.
RESULTS
It has been established that there was an average of 3.3 cases of CNCDs per surveyed person in the 60—65 age group compared to 4.9 in the subjects aged 90 and older. The degree and proportion of the impact of both individual factors and the set of sociomedical factors on the development of CNCDs have been determined as well. A matrix of individual prediction of the probability of CNCDs occurrence, which can be used to manage the risk factors, has been proposed in order to determine the weighted prognostic risk factors for development of CNDCs. Standardized intensity indicator, relative risk and prognostic coefficient of these diseases have been identified. A flow-synthesis algorithm for artificial intelligence on modeling CNCDs in POWA has been created.
CONCLUSION
A program using artificial intelligence for modeling and prediction of CNCDs in POWA creates conditions for the implementation of measures for reduction of the risk of developing this pathology. The proposed program will allow to minimize healthcare costs and improve patients’ quality of life. The application of software with artificial intelligence and constantly updated database in medicine can contribute to the effective implementation of preventive measures for persons older than working age, which will reduce the incidence of CNDCs.
