Abstract
Modern medical diagnostics is impossible without high-tech means of collecting information about the patient's condition, in particular, the biochemical properties of blood and other tissues, physiological signals in the format of time series, and medical images as well. Extensive use of automatic methods of information processing and decision-making based on mathematical models, expert systems, and artificial intelligence is an integral part of the tomorrow’s medicine. Such approaches significantly increase the accuracy of diagnosis and the effectiveness of the prescribed treatment, but this requires the availability of properly structured databases with the results of both successful / unsuccessful treatments, and a complete set of necessary analyses and tests for each patient. This paper discusses the most important components of such database and public health monitoring system. The main issues are the standardization of data format, approaches, methods and laboratory equipment used to unify monitoring, diagnosis and control over the treatment. The importance of using additional physical parameters of blood cells and tissues to increase the efficiency of medical diagnostics with artificial intelligence is shown. The sedimentation curves corresponding to stable normal, stable increased and unstable erythrocyte aggregation rate are given. It is shown that the time to reach the maximum cell sedimentation rate can be calculated on a 2-phase model of blood suspension, indicators of which could be accumulated in the database, which will allow the extraction of additional diagnostic information using novel statistical and mathematical methods. Typical dependences of erythrocyte dielectric constant curves on temperature for oncology patients are given. It is shown that storage in the database the values of dielectric permittivity of red blood cells measured at different temperature and frequencies of electromagnetic fields applied in the dielectrometer, provides significant material for a more detailed study of the patterns of development of various diseases and finding the most sensitive indices for their timely detection.
Publisher
Ivan Kozhedub Kharkiv National Air Force University KNAFU
Reference39 articles.
1. Abdelhak, M., Grostick, S. and Hanken, M.A. (2016), Health Information. E-Book: Management of a Strategic Resource, Elsevier, 800 p.
2. Andreieva, D.N., Vdovichenko, T.V., Kizilova, N.M. and Nikolaiev, A.S. (2020), “Perspektyvni matematychni metody dlja rannjoji diaghnostyky porushenj systemy krovoobighu ljudyny” [Promising mathematical methods for early diagnosis of human circulatory disorders], Bulletin of V. Karazin Kharkiv National University, Ser. “Mathematical Modelling. Information Technology. Automated Control Systems”, No. 45, pp. 4-9. https://doi.org/10.26565/2304-6201-2020-45-01.
3. Kizilova, N.N. and Pakki, D.M. (2020), “Statystychnyj analiz danykh monitorynghu koronarnogho krovotoku dlja ghemodynamichnoji ocinky stupenja stenozu koronarnykh arterij” [Statistical analysis of coronary blood flow monitoring data for hemodynamic assessment of the degree of coronary artery stenosis], Bulletin of V. Karazin Kharkiv National University, Ser. “Mathematical Modelling. Information Technology. Automated Control Systems”, No. 45, pp. 50-55. https://doi.org/10.26565/2304-6201-2020-45-06.
4. Batyuk, L.V. and Kizilova, N.N. (2019), Novel monitoring system for quantitative estimation of efficient medical treatment of diseases based on dielectric properties of blood samples, Bulletin of V. Karazin Kharkiv National University, Series “Mathematical Modelling. Information Technology. Automated Control Systems”, No. 43, pp. 4-10. https://doi.org/10.26565/2304-6201-2019-43-01.
5. Kizilova, N. (2019), Multidisciplinary approaches in cancer diagnosis and treatment: towards patient-specific predictive oncology, AS Cancer Biology, Vol. 3, No. 8, pp. 1-2.