Time Series Prediction of Heart Rate Using Deep Learning Models

Year 2024, Volume: 4 Issue: 2, 79 - 86, 27.12.2024

Emir Evcil

Abstract

Cardiovascular diseases are among the leading causes of mortality worldwide and represent a significant global health burden, affecting millions of individuals each year. Early diagnosis of these diseases is critical not only for improving patient survival rates but also for ensuring the economic sustainability of healthcare systems. Heart rate values serve as essential biological indicators, providing important insights into cardiovascular health and offering potential utility in early diagnosis. In this study, conducted a comprehensive time series analysis to predict the next 5-minute heart rate values based on a 3-minute segment of pulse data collected from healthy individuals. Employed four deep learning models—Recurrent Neural Network (RNN), Bidirectional Long Short-Term Memory (BI-LSTM), Gated Recurrent Unit (GRU), and Long Short-Term Memory (LSTM)—to analyze the constructed dataset. The predictive performances of these models were rigorously compared using the Root Mean Square Error (RMSE) metric, which serves as a reliable measure of accuracy in regression tasks. Findings indicate that deep learning techniques, particularly LSTM and its variants, hold significant promise for enhancing the accuracy of heart rate predictions. This study underscores the potential of these advanced methodologies in the early diagnosis of cardiovascular diseases, aiming to offer new perspectives for the development of clinical decision support systems that could ultimately improve patient outcomes and optimize healthcare delivery.

Keywords

BiLSTM, GRU, Heart Rate Prediction, LSTM, RNN, Time Series

References

• World Health Organization, “Cardiovascular Diseases (CVDs),” Available: https://www.who.int/news-room/fact-sheets/detail/cardiovascular-diseases-(cvds).
• H. Rahman, M. U. Ahmed, and S. Begum, “Vision-based remote heart rate variability monitoring using camera,” in *Internet of Things (IoT) Technologies for HealthCare: 4th Int. Conf., HealthyIoT 2017, Angers, France, Oct. 24-25, 2017, Proc.*, 2018, pp. 10–18.
• A. Esteva, A. Robicquet, B. Ramsundar, V. Kuleshov, M. DePristo, K. Chou, and J. Dean, “A guide to deep learning in healthcare,” *Nat. Med.*, vol. 25, no. 1, pp. 24–29, 2019.
• M. Oyeleye, T. Chen, S. Titarenko, and G. Antoniou, “A predictive analysis of heart rates using machine learning techniques,” *Int. J. Environ. Res. Public Health*, vol. 19, no. 4, pp. 2417, 2022.
• S. Makridakis, E. Spiliotis, and V. Assimakopoulos, “M5 accuracy competition: Results, findings, and conclusions,” *Int. J. Forecast.*, vol. 38, no. 4, pp. 1346–1364, 2022.
• R. J. Hyndman, *Forecasting: Principles and Practice*. OTexts, 2018.
• R. B. Govindan, A. N. Massaro, N. Niforatos, and A. Du Plessis, “Mitigating the effect of non-stationarity in spectral analysis—An application to neonate heart rate analysis,” *Comput. Biol. Med.*, vol. 43, no. 12, pp. 2001–2006, 2013.
• Y. Qin, D. Song, H. Chen, W. Cheng, G. Jiang, and G. W. Cottrell, “A dual-stage attention-based recurrent neural network for time series prediction,” in *Proc. 26th Int. Joint Conf. Artif. Intell.*, 2017, pp. 2627–2633.
• K. Cho, B. van Merrienboer, Ç. Gülçehre, D. Bahdanau, F. Bougares, H. Schwenk, and Y. Bengio, “Learning phrase representations using RNN encoder–decoder for statistical machine translation,” in *Proc. 2014 Conf. Empir. Methods Nat. Lang. Process.*, 2014, pp. 1724–1734.
• D. Bahdanau, K. Cho, and Y. Bengio, “Neural machine translation by jointly learning to align and translate,” *arXiv* preprint arXiv:1409.0473, 2014.
• Y. Liang, S. Ke, J. Zhang, X. Yi, and Y. Zheng, “Geoman: Multi-level attention multi-order for geo-sensory time series prediction,” in *Proc. 27th Int. Joint Conf. Artif. Intell.*, 2018, pp. 3428–3434.
• H. Lin, S. Zhang, Q. Li, Y. Li, J. Li, and Y. Yang, “A new method for heart rate prediction based on LSTM-BiLSTM-Att,” *Measurement*, vol. 207, p. 112384, 2023.
• H. Ni et al., “Time Series Modeling for Heart Rate Prediction: From ARIMA to Transformers,” *arXiv* preprint arXiv:2406.12199, 2024.
• A. Staffini, T. Svensson, U. I. Chung, and A. K. Svensson, “Heart rate modeling and prediction using autoregressive models and deep learning,” *Sensors*, vol. 22, no. 1, p. 34, 2021.
• R. Salles, K. Belloze, F. Porto, P. H. Gonzalez, and E. Ogasawara, “Nonstationary time series transformation methods: An experimental review,” *Knowl.-Based Syst.*, vol. 164, pp. 274–291, 2019.
• J. Bergstra and Y. Bengio, “Random Search for Hyper-Parameter Optimization,” *J. Mach. Learn. Res.*, vol. 13, pp. 281–305, 2012.