Optimizing hospital stays by machine learning and risk analysis
Hospitals have access to a massive amount of data that is already collected for medical and administrative purposes. However, the potential of this data is not fully exploited yet. This stems from the difficulty of extracting useful information and patterns from the scattered and partly inconsistent hospital data. The aim of this research is to find actionable information and patterns that can enhance risk management in hospitals and feed the resulting insights into an intelligent decision-support system according to the different needs of relevant stakeholders. Therefore, different risk factors and stakeholders in the hospital context are analyzed to determine potential areas for improvement. By applying predictive analytics methods, a set of risk prediction models is developed and aggregated to visualize the gathered insights for relevant stakeholders.