Abstract
Effective hospital outpatient forecasting is an important prerequisite for modern hospitals to implement intelligent management of medical resources. As outpatient visits flow may be complex and diverse volatility, we propose a hybrid Autoregressive Integrated Moving Average (ARIMA)-Long Short Term Memory (LSTM) model, which hybridizes the ARIMA model and LSTM model to obtain the linear tendency and nonlinear tendency correspondingly. Instead of the traditional methods that artificially assume the linear components and nonlinear components should be linearly added, we propose employing backpropagation neural networks (BP) to imitate the real relationship between them. The proposed hybrid model is applied to real data analysis and experimental analysis to justify its performance against single ARIMA model, single LSTM model and the hybrid ARIMA-LSTM model based on the traditional method. Compared with competitors, the proposed hybrid model produced the lowest RMSE, MAE and MAPE. It achieves more accurate and stable prediction. Therefore, the proposed model can be a promising alternative in outpatient visit predictive problems.
Similar content being viewed by others
Data availability
The outpatient visit data used to support the findings of this study are available from the corresponding author upon request.
References
Box GEP, Jenkins GM (1976) Time series analysis: forecasting and control. Holden-Day, San Francisco. https://doi.org/10.1080/00401706.1977.10489562
Dechanont S, Jedsadayanmata A, Butthum B, Kongkaew C (2018) Hospital outpatient visits associated with medication-related problems in Thailand: a multicenter prospective observational study. J Patient Saf. https://doi.org/10.1097/PTS.0000000000000367
Flores JJ, Rodriguez H, Graff M (2010) Reducing the search space in evolutive design of ARIMA and ANN models for time series prediction. Adv Soft Comput. https://doi.org/10.1007/978-3-642-16773-7_28
Gers F, Eck D (2010) Applying LSTM to time series predictable through time-window approaches. Int Conf Artif Neural Netw. https://doi.org/10.1007/3-540-44668-0_93
Hadavandi E, Shavandi H, Ghanbari A, Abbasian-Naghneh S (2012) Developing a hybrid artificial intelligence model for outpatient visits forecasting in hospitals. Appl Soft Comput 12:700–711. https://doi.org/10.1016/j.asoc.2011.09.018
Hillmer SC, Tiao GC (1982a) An ARIMA-model-based approach to seasonal adjustment. J Am Stat Assoc 77:63-70S. https://doi.org/10.1080/01621459.1982.10477767
Hillmer SC, Tiao GC (1982b) An ARIMA-model-based approach to seasonal adjustment. J Am Stat Assoc. https://doi.org/10.2307/2287770
Huang J-g, Luo H, Wang H-j, Long B (2009) Prediction of time sequence based on GA-BP neural net. J Univ Electron Sci Technol China 5:029. https://doi.org/10.3969/j.issn.1001-0548.2009.05.028
Janardhanan D, Barrett E (2017) CPU workload forecasting of machines in data centers using LSTM recurrent neural networks and ARIMA models: 55–60 Internet Technology & Secured Transactions. https://doi.org/10.23919/icitst.2017.8356346
Kadri F, Harrou F, Chaabane S, Tahon C (2014) Time series modelling and forecasting of emergency department overcrowding. J Med Syst 38:107. https://doi.org/10.1007/s10916-014-0107-0
Kalpakis K, Gada D, Puttagunta V (2000) Distance measures for effective clustering of ARIMA time-series. IEEE Int Conf Data Min. https://doi.org/10.1109/ICDM.2001.989529
Khazaee PR, Bagherzadeh J, Niazkhani Z, Pirnejad H (2019) Predicting the function of transplanted kidney in long-term care processes: Application of a hybrid model. J Biomed Inform 91:103116. https://doi.org/10.1016/j.jbi.2019.103116
Kumar M, Thenmozhi M (2014) Forecasting stock index returns using ARIMA-SVM, ARIMA-ANN, and ARIMA-random forest hybrid models. Int J Bank Acc Financ 5:284. https://doi.org/10.1504/IJBAAF.2014.064307
Li Y, Wu F, Zheng C, Hou K, Wang K, Sun N (2014) Predictive analysis of outpatient visits to a grade 3, class a hospital using ARIMA model. In: Proceedings of the 2014 International symposium on information technology (ISIT 2014). Dalian: CRC Press, 2015. p 285 https://doi.org/10.1201/b18776-55
Liu C (2009) Price forecast for gold futures based on GA-BP neural network. Int Conf Manage Serv Sci. https://doi.org/10.1109/ICMSS.2009.5302242
Luo L, Zhang X, He X (2017) Hospital daily outpatient visits forecasting using a combinatorial model based on ARIMA and SES models. BMC Health Serv Res 17:469. https://doi.org/10.1186/s12913-017-2407-9
Mitiku E, Amsalu A, Tadesse BT (2018) Pediatric urinary tract infection as a cause of outpatient clinic Visits in Southern Ethiopia: a cross sectional study. Ethiopian J Health Sci 28:187–196. https://doi.org/10.4314/ejhs.v28i2.10
Qian F, Chen L, Li J, Ding C, Chen X, Wang J (2019) Direct prediction of the toxic gas diffusion rule in a real environment based on LSTM. Int J Environ Res Public Health 16:2133. https://doi.org/10.3390/ijerph16122133
Salman AG, Heryadi Y, Abdurahman E, Suparta W (2018) Weather forecasting using merged long short-term memory model (LSTM) and autoregressive integrated moving average (ARIMA) model. J Comput Sci 14:930–938. https://doi.org/10.3844/jcssp.2018.930.938
Siami-Namini S, Tavakoli N, Siami Namin A (2018) A Comparison of ARIMA and LSTM in Forecasting Time Series.2018. 17th IEEE International Conference on Machine Learning and Applications (ICMLA) 1394-1401. https://doi.org/10.1109/icmla.2018.00227
Sun Y, Heng BH, Seow YT, Seow E (2009) Forecasting daily attendances at an emergency department to aid resource planning. BMC Emerg Med 9:1. https://doi.org/10.1186/1471-227X-9-1
Sundaram DSB et al (2017) Improving workload management for reducing readmissions and follow-up visits in outpatient practice using a novel application—Cohort Knowledge Solutions (CKS). IEEE Int Conf Electro Inf Tech. https://doi.org/10.1109/EIT.2017.8053344
Wang S, Phillips P, Liu A (2017) Tea category identification using computer vision and generalized eigenvalue proximal SVM[J]. Fundamenta Informaticae. https://doi.org/10.3233/FI-2017-1495
Wang C, Feng L, Chen K (2019) The impact of ambient particulate matter on hospital outpatient visits for respiratory and circulatory system disease in an urban Chinese population. Sci Total Environ 666:672–679. https://doi.org/10.1016/j.scitotenv.2019.02.256
Xing S, Ju J, Xing J (2018) Research on hot-rolling steel products quality control based on BP neural network inverse model. Neural Comput Appl. https://doi.org/10.1007/s00521-018-3547-5
Yu L, Hang G, Tang L, Zhao Y, Lai KK (2017) Forecasting patient visits to hospitals using a WD&ANN-based decomposition and ensemble model Eurasia. J Math Sci Technol Educ. https://doi.org/10.12973/ejmste/80308
Zhang GP (2003) Time series forecasting using a hybrid ARIMA and neural network model. Neurocomputing 50:159–175. https://doi.org/10.1016/s0925-2312(01)00702-0
Zhang X, Zhang Q, Zhang G, Nie Z, Gui Z, Que H (2018a) A novel hybrid data-driven model for daily land surface temperature forecasting using long short-term memory neural network based on ensemble empirical mode decomposition. Int J Environ Res Public Health 15:1032. https://doi.org/10.3390/ijerph15051032
Zhang Y, Luo L, Yang J, Liu D, Kong R, Feng Y (2018b) A hybrid ARIMA-SVR approach for forecasting emergency patient flow. J Ambient Intell Humaniz Comput. https://doi.org/10.1007/s12652-018-1059-x
Zhang Z, Lu J, Zhou G, Liao X (2018c) Research on tool wear prediction based on LSTM and ARIMA. https://doi.org/10.1145/3297730.3297732
Zhao MJ, Geng XY, Cui LL, Zhou JW, Zhang J (2017) Association between ambient PM(l0)/PM(2.5) concentration and outpatient department visits due to respiratory disease in a hospital in Jinan, 2013–2015: a time series analysis. Zhonghua liu xing bing xue za zhi = Zhonghua liuxingbingxue zazhi 38:374
Zhou L, Xia J, Yu L, Wang Y, Shi Y, Cai S, Nie S (2016) Using a hybrid model to forecast the prevalence of Schistosomiasis in humans. Int J Environ Res Public Health 13:355. https://doi.org/10.3390/ijerph13040355
Acknowledgements
We thank the First Hospital of Shanxi Medical University and the People's Hospital of Shanxi Province for sharing the data needed for this study.
Funding
This research and article was not funded.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare no conflicts of interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
About this article
Cite this article
Deng, Y., Fan, H. & Wu, S. A hybrid ARIMA-LSTM model optimized by BP in the forecast of outpatient visits. J Ambient Intell Human Comput 14, 5517–5527 (2023). https://doi.org/10.1007/s12652-020-02602-x
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12652-020-02602-x