Machine learning methods comparisons to predict LOS performance for Patients undertaking Knee replacement Surgery

Abstract

Knee replacement surgery is a complex orthopedic procedure that often necessitates a prolonged hospitalization period, contributing to elevated medical expenses. The duration of a patient's hospital stays, commonly referred to as the length of stay (LOS), serves as a vital indicator for healthcare providers and hospital administrators to evaluate the overall efficiency of patient care and resource management. Implementing effective measures to minimize hospital stays is crucial for improving bed occupancy rates and reducing financial burdens on healthcare systems. This research aimed to forecast the length of stay (LOS) for individuals undergoing knee replacement surgery at the A.O.R.N. "Antonio Cardarelli" hospital in Naples, Italy. The study focused on comparing the predictive capabilities of machine learning algorithms and neural networks to determine which approach yielded superior performance. The findings presented in this paper highlight the effectiveness of neural networks through an in-depth examination of predictive accuracy and error analysis.

Ähnliche Arbeiten