The kidneys are vital organs that have a significant role in maintaining the body's equilibrium of water by eliminating excess fluids and filtering out harmful substances. Because of this, kidney ailments hold a pivotal spot in the realm of clinical medicine within a hospital setting. It is imperative to thoroughly evaluate a patient's concurrent health conditions and their complete clinical profile, as this is pivotal in comprehending this malady, which could potentially lead to an extension in the patient's hospitalization period. This extension in hospital stays casts its impact across the spectrum of hospital administration, leading to escalated operational expenses. Thus, there exists a compelling interest and necessity to scrutinize the duration of hospital stays (referred to as Length of Stay or LOS) and to introduce models that offer the capability to prognosticate such a parameter. The objective of this research is to scrutinize the Length of Stay for all individuals admitted to the Nephrology Department at the University Hospital “Federico II” located in Naples, Italy, during the timeframe of 2019 to 2021. In this study, the prediction of the Length of Stay was conducted utilizing machine learning algorithms such as Decision Trees (DT), Random Forest (RF), and Naive Bayes (NB). The Random Forest algorithm exhibited a commendable accuracy of 86%, whereas the accuracy values for the remaining algorithms were markedly lower.