Selection of features for modeling the risk of fatal outcomes in patients after myocardial infarction or unstable angina

Aim. To identify main predictors of fatal outcome based on the regional database of patients with myocardial infarction (MI) or unstable angina (UA).

Material and methods. The retrospective study included 1515 patients with UA and MI, which accounted for 55,3% of patients treated in the period 2012-2016. The median follow-up was 62 [36; 71] months. The criterion for a complicated course of coronary artery disease (CAD) is fatal outcome. Group 1 included 238 patients with fatal outcome of CAD, while group 2 — 1277 patients with favorable outcome. All obtained data were structured as an Excel file. Variables available to most medical institutions (except coronary angiography data) were extracted from medical records. For automatic feature extraction, we used an ensemble machine learning algorithm developed by Yandex — CatBoost (Categorical Boosting).

Results. Mortality over 62-month follow-up was 15,4%. The study used 47 quantitative and qualitative (categorical) features. The filter feature selection identified significant quantitative characteristics, including age, left ventricular (LV) ejection fraction (EF), glomerular filtration rate, creatinine, body mass index, height, weight, body surface area (BSA), red blood cells, hemoglobin, glucose, total cholesterol (TC), lowdensity lipoprotein cholesterol, high-density lipoprotein cholesterol, heart rate, LV end-diastolic volume index, LV end-systolic volume index, pulmonary artery systolic pressure. There were following categorical variables: heart failure (HF), heart failure Killip class, old myocardial infarction, comorbidity, chronic kidney disease, angina pectoris, diabetes, atrial fibrillation, positive troponins, S-T deviation, coronary angiography, percutaneous coronary intervention, nosological unit (UA, anterior or inferior MI). An automatic feature selection using a machine learning algorithm identified the following most significant features determining the probability of death: age, LVEF, BSA, creatinine level, systolic blood pressure, HF, comorbidity, nosological unit.

Conclusion. Forty-seven available clinical features were selected from the medical records of patients with MI and UA. There were 8 following most significant parameters for predicting a fatal outcome according to machine selection results: age, LVEF, BSA, creatinine level, systolic blood pressure, HF, comorbidity, nosological unit.

1. Geltser BI, Tsivanyuk MM, Shakhgeldyan KI, еt al. Machine lear¬ning as a tool for diagnostic and prognostic research in coronary artery disease. Russian Journal of Cardiology. 2020; 25(12):3999. (In Russ.) doi:10.15829/1560-4071-2020-3999.

2. Gusev AV, Gavrilov DV, Korsakov IN, et al. Prospects for the use of machine learning methods for predicting cardiovascular disease. Arti¬ficial Intelligence in Healthcare. 2019;3:41-7. (In Russ.)

3. Moshawrab M, Adda M, Bouzouane A, et al. Reviewing Federated Machine Learning and Its Use in Diseases Prediction. Sensors. 2023;23:2112. doi:10.3390/s23042112.

4. Haq AU, Li JP, Memon MH, et al. A Hybrid Intelligent System Fra-mework for the Prediction of Heart Disease Using Machine Learning Algorithms. Mob Inf Syst. 2018;8:1-21. doi:10.1155/2018/3860146.

5. Al-Zaiti SS, Alghwiri AA, Hu X, et al. A clinician’s guide to understanding and critically appraising machine learning studies: a checklist for Ruling Out Bias Using Standard Tools in Machine Learning (ROBUST-ML). Eur Heart J, Digit Health. 2022;3:125-40. doi:10.1093/ehjdh/ztac016.

6. Huang C, Murugiah K, Mahajan S, et al. Enhancing the prediction of acute kidney injury risk after percutaneous coronary intervention using machine learning techniques: A retrospective cohort study. PLoS Med. 2018;15(11):e1002703. doi:10.1371/journal.pmed.1002703.

7. Mirza B, Wang W, Wang J, et al. Machine Learning and Integrative Analysis of Biomedical Big Data. Genes. 2019;10(2):87. doi:10.3390/genes10020087.

8. Johnson KW, Soto JT, Glicksberg BS, et al. Artificial Intelligence in Cardiology. J Am Coll Сardiol. 2018;71(23):2668-79. doi:10.1016/j.jacc.2018.03.521.

9. Fox KAA, Dabbous OH, Goldberg RJ. Prediction of risk of death and myocardial infarction in the six months after presentation with acute coronary syndrome: prospective multinational obser¬vational study (GRACE). Br Med J. 2006;333:1091-4. doi:10.1136/bmj.38985.646481.55.

10. Berns SA, Shmidt EA, Klimenkova AV, et al. Using the GRACE Score to Assess Long-term Risk in Patients with Non-ST Elevation Acute Coronary Syndrome. Doctor.Ru. 2019;2(157):12-8. (In Russ.) doi:10.31550/1727-2378-2019-157-2-12-18.

11. Sedykh DY, Veliyeva RM, Kashtalap VV, et al. Comparison of prog-nostic value of risk assessment scores in patients with myocardial infarction. Complex Issues of Cardiovascular Diseases. 2019; 8(4):46-55. (In Russ.) doi:10.17802/2306-1278-2019-8-4-46-55.

12. Ali BW, Pesaranghader A, Avram R, et al. Implementing Machine Learning in Interventional Cardiology: The Benefits Are Worth the Trouble. Front Cardiovasc Med. 2021;8:711401. doi:10.3389/fcvm.2021.711401.

13. Boytsov SA, Alekyan BG, Shakhnovich RM, et al. What is changing in the treatment of acute coronary syndrome in the Russian Federation? Rational Pharmacotherapy in Cardiology. 2022; 18(6):703-9. (In Russ.) doi:10.20996/1819-6446-2022-12-14.

14. Martsevich SYu, Ginzburg ML, Kutishenko NP, et al. A Lyuberetskiy Study of Mortality Among Patients with Prior Acute Myocardial Infarction: the First Results of the LIS Study. The Clinician. 2011;5(1):24-7. (In Russ.) doi:10.17650/1818-8338-2011-1-24-27.

15. Erlikh AD on behalf of Participants of the RECORD-3 Registry. Relationship Between Degree of Coronary Vascular Bed Involvement and Characteristics of Stenting With Short-Term and Long-Term Outcomes in Patients With Acute Coronary Syndrome (Data of the RECORD-3 Registry). Kardiologiia. 2018;58(5):5-12. (In Russ.) doi:10.18087/cardio.2018.5.10109.

16. Boytsov SA, Shakhnovich RM, Erlikh AD, et al. Registry of Acute Myocardial Infarction. REGION-MI — Russian Registry of Acute Myocardial Infarction. Kardiologiia. 2021;61(6):41-51. (In Russ.) doi:10.18087/cardio.2021.6.n1595.

17. Collet J-Ph, Thiele H, Barbato E, et al. 2020 ESC Guidelines for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation. Russian Journal of Cardiology. 2021;26(3):4418. (In Russ.) doi:10.15829/1560-4071-2021-4418.

18. Staroverov II, Shakhnovich RM, Gilyarov MYu, et al. Eurasian Clinical Gudelines on Diagnosis and Treatment of Acute Coronary Syndrome with ST Segment Elevation (STEMI). Eurasian Heart Journal. 2020; (1):4-77. (In Russ.) doi:10.24411/2076-4766-2020-10001.

19. Erlikh AD. Twelve Months Outcomes in Patients with Acute Co-ronary Syndrome, by the National Registry RECORD-3. Russian Journal of Cardiology. 2018;(3):23-30. (In Russ.) doi:10.15829/1560-4071-2018-3-23-30.

20. Fordyce СВ, Giugliano RP, Cannon CP, et al. Cardiovascular Events and Long-Term Risk of Sudden Death Among Stabilized Patients After Acute Coronary Syndrome: Insights From IMPROVE-IT. J Am Heart Assoc. 2022;11:e022733. doi:10.1161/JAHA.121.022733.

21. Tereshchenko SN, Galyavich AS, Uskach TM, et al. Clinical practice guidelines for Chronic heart failure. Russian Journal of Cardiology. 2020;25(11):4083. (In Russ.) doi:10.15829/1560-4071-2020-4083.

22. Reznik EV, Nikitin IG. Cardiorenal Syndrome in Patients with chronic Heart Failure as a Stage of the Cardiorenal Continuum (Part 1): Definition, Classifiсation, Pathogenesis, Diagnosis, Epydemiology. The Russian Archives of Internal Medicine. 2019;9(1):5-22. (In Russ.) doi:10.20514/2226-6704-2019-9-1-5-22.

23. Khudainetova LA, Efimova LP, Mirzalieva MN. Internation of the Charlson Comorbidity Index with Number of Rehospitalizations of Comorbid Cardiological Patients. Vestnik SurGU. Meditsina. 2022;2(52):14-21. (In Russ.) doi:10.34822/2304-9448-2022-2-14-21.

24. Masci PG, Ganame J, Francone M, et al. Relationship between location and size of myocardial infarction and their reciprocal influences on post-infarction left ventricular remodelling. Eur Heart J. 2011;32:1640-8. doi:10.1093/eurheartj/ehr064.

25. Hermansson J, Bøggild H, Hallqvist J, et al. Interaction between Shift Work and Established Coronary Risk Factors. Int J Occup Environ Med. 2019;10(2):57-65. doi:10.15171/ijoem.2019.1466.

26. Merlo J, Mulinari S, Wemrell M, et al. The tyranny of the averages and the indiscriminate use of risk factors in public health: The case of coronary heart disease. SSM Popul Health. 2017;3:684-98. doi:10.1016/j.ssmph.2017.08.005.

27. Mok Y, Sang Y, Ballew SH, et al. American Heart Association’s Life’s Simple 7 at Middle Age and Prognosis After Myocardial Infarction in Later Life. J Am Heart Assoc. 2018;7:e007658. doi:10.1161/JAHA.117.007658.

28. Kanenawa К, Yamaji K, Kohsaka S, et al. Age-Stratified Preva¬lence and Relative Prognostic Significance of Traditional Atherosclerotic Risk Factors: A Report from the Nationwide Registry of Percutaneous Coronary Interventions in Japan. J Am Heart Assoc. 2023;12:e030881. doi:10.1161/JAHA.123.030881.

29. Patel RS, Schmidt AF, TraganteV, et al. Association of Chromo¬so¬me 9p21 With Subsequent Coronary Heart Disease Events A GENIUS-CHD Study of Individual Participant. Circ, Genom Precis Med. 2019;12(4):е002471. doi:10.1161/CIRCGEN.119.002471.

30. Simonetto С, Heier M, Peters A, et al. From Atherosclerosis to Myocardial Infarction: A Process-Oriented Model Investigating the Role of Risk Factors. Am J Epidemiol. 2022;191(10):1766-75. doi:10.1093/aje/kwac038.