Method of thromboembolism prediction in advanced atherosclerosis

Aim. To develop a method for thromboembolism prediction in patients with ad­vanced atherosclerosis.

Material and methods. The study involved 170 patients with advanced atheroscle­rosis and 110 patients with local atherosclerotic lesions. At the first stage, speciali­zed experts generated signs characterizing the severity of prothrombotic conditions. To determine whether patients belonged to the first or second group, the multivariate statistics method was used. Decision rules (DRs) was developed based on soft computing (SC), as well as the methodology for the synthesis of hybrid fuzzy DRs.

Results. According to the discriminant function coefficient (DFC), the division into classes was made with the formation of an intersection area, which affected the diagnostic sensitivity of this method (=86%). Due to difficulties of separating more heterogeneous groups with a small sample, RP was synthesized according to the atherosclerosis severity classification using SC technology. Atherosclerosis severity classification was adopted according to fuzzy DRs: initiation stage (up to 20%); reverse stage (21-40%); progressive stage (41-65%); critical stage (more than 65%). As­ses­sment of thromboembolism risk in advanced atherosclerosis in the examined pati­ents showed the following trends: progressive stage — 55% of patients; critical stage — 21% of patients; reverse stage — 16% of patients; initiation stage — 7% of patients.

Conclusion. DFC for DRs makes it possible to identify groups with high and low thromboembolism risk. The resulting final fuzzy DRs make it possible to differentiate the atherothrombotic condition in advanced atherosclerosis by severity, which can help to timely determine preventive and therapeutic measures.

1. Klimchuk AV, Beloglazov VA, Zayaeva AA. Atherosclerosis: immunological aspects of patho­genesis, the role of inflammation, therapeutic strategies, prospects for the use of nanotechnology. Tavricheskiy mediko-­biologicheskiy vestnik. 2021;3:77-89. (In Russ.) doi:10.37279/2070-8092-2021-24-3-77-89.

2. Meldekhanov TT, Yesergepova SR, Pirzhanov BT, et al. To the pathogenesis and morphogenesis of atherosclerosis. Actual Problems of Theoretical and Clinical Medi­cine. 2021;4:29-7. (In Russ.) doi:10.24412/2790-1289-2021-42937.

3. Aimagambetova AO. Аtherogenesis and inflammation. Science & Healthcare. 2016; 1:24-39. (In Russ.) doi:10.34689/SH.2016.18.1.002.

4. Fatenkov OV, Simerzin VV, Nizametdinova DR, et al. Diagnosis of multifocal atherosclero­sis in low-to moderate-risk patients and their restratification. Bulletin of the Medical Insti­tute "REAVIZ": rehabilitation, doctor and health. 2020;1(43):17-26. (In Russ.) EDN: OKCTOO.

5. Komarov AL, Novikova ЕS, Guskova EV, et al. New possibilities of antithrombotic therapy of patients with peripheral and widespread atherosclerotic lesion. Rational pharma­cotherapy in cardiology. 2018;14(2):272-83. (In Russ.) doi:10.20996/1819-6446-2018-14-2-272-283.

6. Korenevskiy NA. Application of fuzzy logic for decision-­making in medical expert sys­tems. Biomedical Engineering. 2015;1:33-5. (In Russ.) EDN: TQMZCZ.

7. Korenevskiy NA, Rodionova SN, Khripina II. Methodology of synthesis of hybrid fuzzy decision rules for medical intelligent decision support systems. Staryj Oskol: TNT, 2019. p. 472. (In Russ.) ISBN: 978-5-94178-602-2.

8. Korenevsky NA, Artemenko MV, Provotorov VYa, et al. Method of fuzzy synthesis decision rule based on a model system interrelation for solving problems of prediction and diagnosis of diseases. System analysis and management in biomedical systems. 2014;13(4):881-6. (In Russ.) EDN: TCWKLF.

9. Korenevskiy NA, Serebrovskiy VV, Razumova KV, et al. The synthesis method of hybrid fuzzy decision-­making models for the state assessment and biotechnology systems control. Biomedicine Radioengineering. 2016;9:68-74. (In Russ.) EDN: XQOLJV.

10. Korenevskiy NA, Shutkin AN, Boytsova EA, et al. Assessment and management of health status based on G. Rasch models. Biomedical Engineering. 2015;6:37-40. (In Russ.) EDN: VCFQKP.

11. Korenevskiy NA, Shutkin AN, Gorbatenko SA, et al. Assessment and management of students' health status (based on hybrid intelligent technologies). Staryj Oskol: TNT, 2016. p. 472. (In Russ.) ISBN: 978-5-94178-504-9.

12. Korenevsky NA, Shutkin AN, Boytsova EA, et al. Classification and measurement of the level of functional states on the basis of fuzzy modification measurement theory of latent variables. Biomedicine Radioengineering. 2016;3:53-60. (In Russ.) EDN: WAGHLN.

13. Korenevsky NA, Pozin AO, Starodubceva LV, et al. The role of exploratory analysis in the synthesis of fuzzy decision rules for assessing the state of biotechnical systems. Vestnik nauchnyh konferencij. 2016;(9-2(13):51-4. (In Russ.) EDN: UROXKD.