Deep Learning-Based Patient Visits Forecasting Using Long Short Term Memory

摘要

Predicting the number of patient visits to hospitals has been acknowledged remarkably helpful in the decision-making related to allocating limited human and material resources of hospitals. More accurate its prediction can contribute to higher efficiency of hospital management without any doubt. This study proposes Long Short Term Memory (LSTM) method to predict the number of patient visits to a community health center in South Tangerang City on a monthly basis. LSTM is chosen because it is known to perform well in time series data. Toward a fair evaluation, the performance of the proposed model is compared to the AutoRegressive Integrated Moving Average (ARIMA), simple exponential smoothing, linear regression, and conventional artificial neural networks. The results show that the proposed LSTM model surpasses the benchmark methods with a Mean Absolute Percentage Error (MAPE) value of 4.714.