The article presents a theoretical overview of International and domestic experience in improving statistical methods of data processing in medical universities with the help of machine learning. Since statistical processing of the results of biomedical experiments and data from daily medical practice is complex, the establishment and validity of new scientific hypotheses in the study of a patient's condition or disease as a diagnostic goal are statistical programs based on methods and stages of statistical analysis. For students of medical education, as future specialists in professional practice, the study of biostatistics is intended for the analysis of experimental data and observations, if the main method of statistics is abstraction, then the treatment of a particular disease is considered in medicine. Data collection and control using statistical technologies is the basis of a specific disease.