Рақамли технологияларнинг назарий ва амалий масалалари Том 8 № 4 (2025) · с. 48-57
Assessment of Myocardial Infarction Patient Survival using Machine Learning Methods
Пекось, О.А.
Аннотация
High mortality in the acute and post-hospital periods of myocardial infarction underscores the relevance of accurately stratifying patient risk and assessing survival outcomes. This study compares two complementary approaches: the Cox regression model, including time-dependent covariates, and an ensemble random forest method for binary classification and survival analysis. It is shown that accounting for the dynamics of clinical and laboratory indicators (blood pressure, heart rate, necrosis markers, etc.) improves both the discrimination and calibration of prognostic models. Variants of the instantaneous hazard equation, partial likelihood formulation, and formulas for estimating survival and cumulative hazard are presented. Illustrative examples based on a mixed dataset of real and synthetic data demonstrate the advantages of the combined approach in the presence of censoring and heterogeneous risk profiles. The results confirm the feasibility of integrating such models into the clinical workflow to personalize treatment and rehabilitation strategies after myocardial infarction.
машинное обучениеанализ выживаемостипрогнозирование летальностимодель Коксаслучайный лес выживанияmachine learningsurvival analysismortality predictionCox modelsurvival random forest
Источник метаданных: OAI-PMH архив журнала · Sindex не хранит полный текст, а даёт ссылку на источник.