Estimating the number of emergency patients with artificial intelligence: Bingöl State Hospital example

The study aims to evaluate artificial intelligence-based demand forecasting models in order to more effectively predict the number of patients that may arise in emergency services in the healthcare sector. Focusing on the applicability and effectiveness of this technology, the potential of providing more rational and faster planning and resource management in the healthcare sector was examined with the sample application to be carried out at Bingöl State Hospital. In the study, thirteen independent variables were selected and the total number of patients applying to the emergency department was determined as the dependent variable. Various prediction models were used to predict emergency department admissions to Bingöl State Hospital between 2013 and 2023. Among the time series models, the SARIMA(1,1,1) model showed the highest performance. According to the prediction results with linear regression, low MSE (0.017211) and low RMSE (0.13119) values were obtained. The ANN model made the most accurate predictions with the lowest MSE (0.012172) and RMSE (0.11040) values. It has been shown that among the independent variables, factors such as total population, male population, female population and literacy rate have significant effects on emergency department admissions. It is aimed to increase the quality of health services with the predictions made by taking local demographic and socioeconomic factors into account. Results obtained through the use of ANN and other machine learning algorithms can provide important guidance to healthcare managers in predicting future patient flow and planning resource management more effectively. Predictions of the number of patients for 2024-2025 were calculated using the ANN model and can play an important role in making strategic decisions such as emergency health services planning and resource allocation. Estimated patient numbers constitute a critical source of information for planning and managing hospital capacity. The results highlight the importance of the use of artificial intelligence and data analytics techniques in the healthcare sector and show that they have the potential to make emergency healthcare more effective and efficient.