|
The last decades have been marked by significant progress in the field of pediatric cardiac surgery in the Russian Federation, but the issues of risk assessment of surgical interventions remain relevant and poorly understood. Given the complexity and versatility of such operations, it is necessary to develop a comprehensive risk assessment model that can adequately predict the likelihood of complications in intra- and postoperative periods. Materials and methods. The analysis of medical data of patients who underwent cardiac surgery aged 0 to 18 years in the period from 2015 to 2021, using methods of mathematical statistics and machine learning, was carried out. Among the variables used were the gender and age of the patient, the type of operation, the presence of concomitant diseases, the duration of the operation and the period of hospitalization. Results.
The developed model takes into account variables such as the degree of complexity of the operation (OS), determined based on the analysis of operating protocols and intraoperative data, including the duration of anesthesia (YES) and the amount of blood loss (OK), as well as demographic data of the patient: age (B) and gender (N), and body mass index (BMI), reflecting the physiological state of the body. Particular attention is paid to the risk of developing postoperative organ failure (PNO), the index of which (IPNO) correlates with the total duration of stay in the intensive care unit (ICU).
Keywords:pediatric cardiac surgery, risk assessment, intraoperative data, postoperative complications, medical statistics, machine learning
|