Application of PSO-based LSTM Neural Network for Outpatient Volume Prediction.

Area of Science:

• Artificial Intelligence
• Machine Learning
• Healthcare Management

Background:

• Accurate prediction of hospital outpatient volume is crucial for effective resource allocation and management.
• Traditional forecasting methods may not fully capture the complex temporal dynamics of patient flow.

Purpose of the Study:

• To develop and evaluate a novel outpatient volume prediction model using a long- and short-term memory neural network (LSTM) optimized with particle swarm optimization (PSO).
• To assess the performance improvement of the PSO-optimized LSTM model in predicting hospital outpatient volume compared to an unoptimized model.

Main Methods:

• Historical outpatient volume data from hospital departments was collected.
• A long- and short-term memory neural network model was designed and implemented.
• Particle swarm optimization (PSO) algorithm was employed to fine-tune the parameters of the LSTM network.

Main Results:

• The PSO-optimized LSTM model demonstrated a significant reduction in Root Mean Square Error (RMSE) by 48.5% on the test set compared to the unoptimized model.
• The optimized model exhibited enhanced accuracy in predicting outpatient volume trends.
• Experimental results confirmed the efficacy of PSO in improving LSTM prediction capabilities.

Conclusions:

• Particle swarm optimization effectively enhances the performance of LSTM networks for outpatient volume prediction.
• The developed PSO-optimized LSTM model provides a valuable tool for data-driven decision support in hospital outpatient management.
• This approach offers a more accurate and reliable method for forecasting patient flow, aiding medical staff in operational planning.