An open access journal
Research and Application of Outpatient Revenue Prediction Model Based on LSTM and BERT in Smart Healthcare
Abstract
With the rapid advancement of information technology, smart healthcare has become a crucial tool for improving the quality and efficiency of medical services. This study analyzes and predicts outpatient revenue based on time series forecasting models using medical data from a hospital's outpatient department. The objective is to optimize resource allocation and enhance hospital management efficiency. For the first and second rehabilitation wards, Long Short-Term Memory (LSTM) and BERT models were constructed to predict outpatient revenue. The experimental results indicate that the BERT model outperforms in capturing global features of the time series. Subsequently, a stacked ensemble method was employed, combining the predictions from the LSTM and BERT models with XGBoost for final prediction, successfully forecasting the outpatient revenue for the third rehabilitation ward. Validation results show that the stacked ensemble model achieved the best performance in outpatient revenue prediction, providing effective support for revenue management in smart healthcare. This paper demonstrates the potential of new productivity in the healthcare field and offers valuable insights for the further development of smart healthcare.
Share and Cite
Article Metrics
References
- Tang Y X ,Yang W W ,Liu Z , et al. Deep learning performance prediction for solar-thermal-driven hydrogen production membrane reactor via bayesian optimized LSTM [J]. International Journal of Hydrogen Energy, 2024, 82 1402-1412.
- Ma S ,Ding W ,Zheng Y , et al. Edge-cloud collaboration-driven predictive planning based on LSTM-attention for wastewater treatment [J]. Computers & Industrial Engineering, 2024, 195 110425-110425.
- Manohar B ,Das R ,Lakshmi M . A hybridized LSTM-ANN-RSA based deep learning models for prediction of COVID-19 cases in Eastern European countries [J]. Expert Systems With Applications, 2024, 256 124977-124977.
- Zheng M ,Luo X . Joint estimation of State of Charge (SOC) and State of Health (SOH) for lithium ion batteries using Support Vector Machine (SVM), Convolutional Neural Network (CNN) and Long Sort Term Memory Network (LSTM) models [J]. International Journal of Electrochemical Science, 2024, 19 (9): 100747-100747.
- Haider F S ,Shah M ,Alarifi S N , et al. The 2023 Mw 6.8 Morocco earthquake induced atmospheric and ionospheric anomalies [J]. Journal of Atmospheric and Solar-Terrestrial Physics, 2024, 262 106323-106323.
- Wang M ,Li B ,Dai G , et al. A dynamic multi-objective optimization algorithm with a dual mechanism based on prediction and archive [J]. Swarm and Evolutionary Computation, 2024, 90 101693-101693.
- Khan K M ,Houran A M ,Kauhaniemi K , et al. Efficient state of charge estimation of lithium-ion batteries in electric vehicles using evolutionary intelligence-assisted GLA–CNN–Bi-LSTM deep learning model [J]. Heliyon, 2024, 10 (15): e35183-e35183.
- Yang D ,Li M ,Guo E J , et al. An attention-based multi-input LSTM with sliding window-based two-stage decomposition for wind speed forecasting [J]. Applied Energy, 2024, 375 124057-124057.
- Dhiman P ,Kaur A ,Gupta D , et al. GBERT: A hybrid deep learning model based on GPT-BERT for fake news detection [J]. Heliyon, 2024, 10 (16): e35865-e35865.
- Panoutsopoulos H ,Garcia E B ,Raaijmakers S , et al. Investigating the effect of different fine-tuning configuration scenarios on agricultural term extraction using BERT [J]. Computers and Electronics in Agriculture, 2024, 225 109268-109268.
- Dharrao D ,MR A ,Mital R , et al. An efficient method for disaster tweets classification using gradient-based optimized convolutional neural networks with BERT embeddings [J]. MethodsX, 2024, 13 102843-102843.
- Hamed K S ,Aziz A J M ,Yaakub R M . Enhanced Feature Representation for Multimodal Fake News Detection Using Localized Fine-Tuning of Improved BERT and VGG-19 Models [J]. Arabian Journal for Science and Engineering, 2024, (prepublish): 1-17.