Machine Learning Algorithms for Hospital Length of Stay Prediction

Efficient prediction of hospital length of stay is crucial for optimizing healthcare operations, resource allocation, and cost-effectiveness. This research harnesses artificial intelligence, specifically machine learning techniques, to enhance the accuracy of predicting hospital Length of Stay. The principal objective is to assist healthcare professionals in selecting the most suitable prediction method based on their specific requirements and constraints. Customized machine learning models have demonstrated success in achieving precise predictions, streamlining the acquisition of patient data, and evaluating both hospitalization and treatment durations. After evaluating various algorithms, the Random Forest algorithm emerged as the most effective for predicting hospital stay durations, yielding a Root Mean Square Error (RMSE) of 0.24 and a coefficient of determination (R2) of 0.94. These compelling findings underscore the efficacy of the Random Forest algorithm in accurately forecasting the duration of hospital stays.