Title: RNN-based deep-learning approach to forecasting hospital system demands: application to an emergency department
Authors: Farid Kadri; Kahina Abdennbi
Addresses: Aeroline and Customer Services, Agence 1024, Sopra Steria Group, 31770, Colomiers, France ' Municipal Health and Hygiene Service (SCHS), Mairie de Montauban, Direction Générale de la Solidarité et de l'Action Sociale, 9 rue de l'Hôtel de ville, 82000, Montauban, France
Abstract: In recent years the management of patient flow is one of the main challenges faced by many hospital establishments, in particular emergency departments (EDs). Increasing number of ED demands may lead to ED overcrowding. One approach to alleviate such problems is to predict patient attendances in order to help ED managers to make suitable decisions. Existing regression and time series such as ARIMA models are mainly linear and cannot describe the stochastic and non-linear nature of time series data. In recent years, recurrent neural networks (RNNs) have been applied as novel alternatives for prediction in various domains. In this paper we propose an RNN deep learning based approach for predicting ED demands. The experiments were carried out on a real database collected from the pediatric emergency department (PED) in Lille regional hospital center, France. The RNN-based deep learning approach was shown to provide a useful tool for predicting ED admissions.
Keywords: ED demands; management of patient flow; ED overcrowding; forecasting ED demands; prediction models; machine learning and deep learning; RNNs; recurrent neural networks; RNN-LSTM; RNN-GRU.
International Journal of Data Science, 2020 Vol.5 No.1, pp.1 - 25
Received: 13 Jan 2020
Accepted: 05 May 2020
Published online: 25 Aug 2020 *