Enhancing Healthcare Operations with Predictive Length of Stay Models
Abstract: Hospital Length of Stay (LOS) predictions offer a proactive approach for managing healthcare operations, allowing providers to allocate resources efficiently, reduce costs, and improve patient outcomes. With daily costs averaging $2,883 and extended stays placing financial strain on facilities, LOS prediction models can transform healthcare operations. This paper evaluates four advanced predictive models - Random Forest, Gradient Boosting, Support Vector Machines (SVM), and Neural Networks - and introduces a hybrid approach combining these models for enhanced accuracy. By comparing the accuracy of each model in LOS prediction and discussing their broader impact on patient care, this paper provides insights into optimal model selection for achieving high reliability in hospital settings.
Keywords: hospital length of stay, healthcare operations, predictive models, patient outcomes, resource allocation
How to Cite?: Umamaheswara Reddy Kudumula, "Enhancing Healthcare Operations with Predictive Length of Stay Models", Volume 13 Issue 7, July 2024, International Journal of Science and Research (IJSR), Pages: 792-796, https://www.ijsr.net/getabstract.php?paperid=SR24715093831, DOI: https://dx.doi.org/10.21275/SR24715093831