The Capabilities of Artificial Intelligence to Simulate the Emergence and Development of Diseases, Optimize Prevention and Treatment Thereof, and Identify New Medical Knowledge
Objectives: To show and clearly demonstrate by examples that the possibilities of the artificial intelligence methods in the field of medicine are much wider than those currently used. Almost all the known studies in this field are reduced to diagnosing various kinds of diseases at a given time, or to predicting their outcomes in an indefinite future. Modern literature lacks information on dynamic mathematical models that treat diseases as time-evolving processes. In this regard, the mathematical modeling method is practically not used to solve the problem of choosing optimal strategies for the prevention and treatment of diseases, which is very important for medical practice. Methods: Four methods of creating dynamic mathematical models based on neural networks were proposed and tested in this paper. The first of these methods implements the idea of adding neural network knowledge with the knowledge embedded in the SCORE scale. The second and third methods represent modifications to the sliding window method. The fourth method has no rigorous justification and refers to the heuristic techniques that reflect the authors' experience in the application of neural network technologies in various fields of knowledge. Results and Conclusion: Based on the cardiovascular system diseases, it has been shown that our mathematical models allow us not only to diagnose diseases at the current moment, but also to predict their appearance and development in future periods of life, and also to select the optimal strategy for their prevention and treatment, taking into account the patient's individual parameters. The article shows the possibility of identifying new medical knowledge using mathematical models. The conclusion that recommendations for the prevention and treatment of cardiac patients should be given strictly individually, taking into account the physiological characteristics of the patient's body, has been made. While for some patients such recommendations as: "to limit the use of sweet", "to stop smoking", "to take drugs that reduce blood pressure," etc. are really useful, for other they can cause harm. The proposed diagnostic and prognostic system allows identifying such non-standard patients and avoiding erroneous recommendations. The demonstration prototype of the diagnostic and prognostic system is freely available in the "Projects" section at the website www.PermAi.ru.