Operation Duration Estimation by Ensemble Methods of Decision Trees

With the multiplied data flow and information accumulation in medical organizations, new opportunities arise to process data and extract meaningful information from them for both therapeutic and organizational purposes. The peculiarity of medical data is the contradiction between the obvious cause-and-effect relationships between influencing factors and target indicators and, often, the complete lack of opportunities to build functional dependencies between them. In this case, application of machine learning methods and algorithms can be a solution to this contradiction. The article considers the task of predicting the duration of surgical operations according to a given set of influencing factors for the purposes of optimal loading of operating units of a medical organization. The data for the study was provided by one of the leading clinics in its field in St. Petersburg.