Development of an algorithm for diagnosing endothelial dysfunction in patients with chronic rhinosinusitis

Purpose of research . The problem of workload of doctors with an increase in the volume of informatization in medicine can be solved through the development of diagnostic algorithms. The integral approach in the work of a doctor includes not only examining the patient, collecting anamnesis, but also taking into account the results of clinical laboratory and radiation research methods. The purpose of the research is to improve the algorithm for diagnosing a patient with chronic rhinosinusitis using a mathematical model. Methods . When compiling the examination algorithm, the requirements of GOST 19.701-90 were applied. Logistic regression was used as a mathematical method for constructing a mathematical model. Correlation analysis and ROC analysis were also used to build a mathematical model. Results. The results of the entire complex of examination of the patient should be entered into the medical information system and integrated into the medical decision support system. According to the results of the calculation using a mathematical model, the patient is diagnosed with the presence or absence of clinical and laboratory manifestations of endothelial dysfunction. Depending on the result obtained, the tactics of further treatment of the patient are built. Conclusion . An algorithm for examining a patient with chronic rhinosinusitis using a mathematical model is proposed. This approach reduces the time for processing the received data for the doctor and making a decision on further patient management tactics. The developed algorithm can be used not only for diagnostic purposes, but also for monitoring the treatment of patients with chronic rhinosinusitis. The use of the algorithm can be recommended to doctors of poly- clinics and hospitals. This tactic of examining the patient reduces the processing time of the received data for the doctor and making a decision on further patient management tactics. The mathematical model does not provide an accurate diagnosis, but only estimates the likelihood of the disease. Various clinical data are also needed to make a diagnosis.