Study of hospitalization for retinal surgery using classification algorithms

The objective of this work is to analyze LOS for patients admitted to the ophthalmology department. The procedures that are performed in this department are many, but our job is to focus attention on retinal diseases. Postoperative time and timely diagnosis are essential for proper patient care. A prolonged length of stay (LOS) average hospital can not only have negative effects on the patient's quality of life but has The LOS is one the main factors affecting on the effectiveness of national health system, which is influence by different contextual information (i.e., inadequate resource management, bed availability and surgical interventions). It is interesting and important to analyze the Length of Stay (LOS) and to implement models that allow to predict such indicator. LOS is related to several variables that describe the patient's condition that affect the management of resources, beds, and the entire department. In this work, we design different Machine Learning models with the aim to predict the LOS. In particular, we evaluate the effectiveness of our approach by analyzing information about 450 patients frmo the University Hospital "Federico II"in Naples (Italy).