Artificial intelligence technologies in predicting genetic disorders and in personalized primary and secondary prevention of brain infarction in young people

Background. Over the past decades, the programs for the prediction and prevention of prevailing cardiovascular diseases have been intensively developing. Up-to-date molecular genetic methods provide scientists with new prospects for the diagnosis, prediction of the outcome and optimal treatment of acute cerebral circulation disorders. Hereditary thrombophilia can be considered as a trigger of ischemic stroke, since in some patients, occlusion of the cerebral arteries due to intravascular thrombosis is revealed during the examination. Knowledge on genetic predisposition of the patient to ischemic stroke will allow us to develop the methods of individualized primary and secondary prevention of the pathology.

Objective. The purpose of the study was to develop a prognostic model based on the design equation of the coefficients of the pathological alleles presence in the genes controlling predisposition to IS according to a set of biochemical indicators. Materials and methods. The genetic, clinical and laboratory results of the examination of 280 people have been analyzed. Group I consisted of patients with IS (n = 180) aged 22 to 45 years (mean age 33.4 Ѓ} 6.57, including 38 patients who experienced recurrent ischemic stroke). Group II included patients with IS (n = 50) aged 52 to 100 years (mean age 73.4 Ѓ} 8.24 years). The control group – group III, consisted of apparently healthy individuals (n = 50) aged 20 to 43 years (average age 31.5 Ѓ} 5.82 years). All patients underwent computed tomography of the brain, ultrasound examination of the brachiocephalic arteries, and echocardiography. Pharmacogenetic investigations as well as venous blood tests were once performed in all the subjects to reveal a genetic predisposition to thrombophilia. Multiple regression analysis (ANOVA) has been used to calculate the prediction coefficients for the presence of pathological alleles.

Results. A mathematical model has been developed at the level of the following genes: angiotensin IIAGTR1 receptor (A1166C), G-protein beta 2 GNB2 (C825T) controlling blood pressure, interleukin-6 IL-6 gene (G-174C) controlling immune response, methionine synthase MTR (A2756G) genes, methylenetetrahydrofolate reductase MTHFR (A1298C), methylenetetrahydrofolate reductase MTHFR (C677T), controlling the level of homocysteine, inhibitor of plasminogen activator PAI-1 (5G/4G), controlling the hemostasis system, platelet receptor fibrinogen GP III a (HPA1-1 a/1 b), controlling aspirin resistance. Calculations of the equation are based on the relationship between the alleles of a particular gene and 22 independent variables. The model is designed to predict the possible presence of genetic thrombophilia.

Conclusion. Thus, it is possible to make recommendations based on the results of standard biochemical studies that allow us to assume the presence of mutations in one of the genes and perform an adjusting genetic assessment. The initial examination of patients with BI can play a principal role in the early identification of the factors that prognostically influence the pathology development. The designed programme can be an effective tool in making clinical decisions for the hospitalized BI population.

1. Oganov R. G. Therapeutic archive. 2004. 2004; 6 (76): 22-24. (In Russ.)

2. Meschia J. F. New Information on the Genetics of Stroke. CurrNeurolNeurosci Rep. 2011; 11 (1): 35-41.

3. Hart R. G., Diener H. C., Connolly S. J. Embolic strokes of undetermined source: support for a new clinical construct – authors’ reply. Lancet Neurol. 2014; 13 (10): 967. DOI: 10.1016/S1474-4422(14)70197-8. PMID: 25231516.

4. Dobrynina L. A., Kalashnikova L. A., Patrusheva N. L., etc. Polymorphism of 5,10-methylenetetrahydrofolate reductase, prothrombin, and blood clotting factor V genes in young patients with ischemic stroke. Clinical medicine. 2012. 2012; 3: 37-40. (In Russ.)

5. Egeberg O. Inherited antithrombin deficiency causing thrombophilia. Thromb. Diath. Haemorgh. 1965; 13: 516-530. PMID: 14347873.

6. Shishkova V. N., Adasheva T. V., Remennik A. Yu., Valyaeva V. N., Shklovsky V. M. Prognostic significance of clinical-anthropometric, biochemical, metabolic, vascular-inflammatory and molecular genetic markers in the development of the first ischemic stroke. Journal of Neurology and Psychiatry named after C.C. Korsakov. 2018; 118 (2): 4-11. (In Russ.) https://doi.org/10.17116/jnevro2018118214-11.

7. Khanr S., Dickerman J. D. Hereditary thrombophilia. Thrombosis Journal. 2006; 4: 15. DOI: 10.1186/1477-9560-4-15.

8. Negrini S. et al. The antiphospholipid syndrome: from pathophysiology to treatment.Clin. Exp. Med. 2017; 3 (17): 257–267.

9. Erkan D. et al. Antiphospholipid Syndrome: Current Research Highlights and Clinical Insights. Springer-Verlag GmbH, 2017. 371 р.

10. Wang C., Dai J., Sun Y., Xie L., Pan L., Hu Z., Shen H. [Genetic risk score: principle, methods and application]. Zhonghua Liu Xing Bing Xue Za Zhi. 2015; 36 (10): 1062-1064. Chinese. PMID: 26837344.

11. Salas E., Farm M., Pich S., Onelöv L., Guillen K., Ortega I., Antovic J. P., Soria J. M. Predictive Ability of a Clinical-Genetic Risk Score for Venous Thromboembolism in Northern and Southern European Populations. TH Open. 2021; 5 (3): e303-e311. DOI: 10.1055/s-0041-1729626. PMID: 34263111; PMCID: PMC8266419.

12. Ramírez Cervantes K. L., Mora E., Campillo Morales S., Huerta Álvarez C., Marcos Neira P., Nanwani Nanwani K. L., Serrano Lázaro A., Silva Obregón J. A., Quintana Díaz M. A Clinical Prediction Rule for Thrombosis in Critically Ill COVID-19 Patients: Step 1 Results of the Thromcco Study. J Clin Med. 2023; 12 (4): 1253. DOI: 10.3390/jcm12041253. PMID: 36835788; PMCID: PMC9966844.

13. Dvorzhak V. S., Shulaev A. V., Ivanovskaya E. A. The experience of using neural networks to predict the outcomes of ischemic stroke. Literature review. Population health and habitat. 2021; 7 (29): 82-85. (In Russ.) DOI: https://doi.org/10.35627/2219-5238/2021-29-7-82-85.

14. Murray N. M., Unberath M., Hager G. D., Hui F. K. Artificial intelligence to diagnose ischemic stroke and identify large vessel occlusions: a systematic review. J Neurointerv Surg. 2020; 12 (2): 156-164. DOI: 10.1136/neurintsurg-2019-015135.

15. Soun J. E., Chow D. S., Nagamine M., et al. Artificial intelligence and acute stroke imaging. AJNR Am J Neuroradiol. 2021; 42 (1): 2-11. DOI: 10.3174/ajnr.A6883.