Deep Learning-Based Time Series Analysis for Early Sepsis Prediction in Intensive Care Units
DOI:
https://doi.org/10.31838/NJAP/08.02.20Keywords:
DL, Intensive Care Units, Time Series Analysis, performanceAbstract
Sepsis, a potentially lethal condition caused by an uncontrolled immune response to an infection, is one of the primary causes of in-hospital mortality. This paper proposes a deep learning-based time series architecture for early and accurate sepsis prediction using continuous physiological inputs and electronic health records (EHRs).The model combines Long Short-Term Memory (LSTM) and Bidirectional Gated Recurrent Units (Bi-GRU) networks to capture temporal relationships in multivariate clinical data, such as heart rate, temperature, respiration rate, blood pressure, and laboratory indicators. Data preprocessing includes normalization, outlier removal, and temporal alignment to ensure high-quality input sequences. Feature extraction is optimized using a time series analysis mechanism that enabling the model to focus on critical time intervals preceding sepsis onset. With an accuracy of 97.45%, sensitivity of 97.15%, specificity of 97.82%, and an AUC of 97.68%, experimental evaluation on benchmark ICU datasets like MIMIC-III shows exceptional predictive performance.Through prompt sepsis response in intensive care units, the suggested architecture offers an efficient early warning system that improves clinical decision-making and patient outcomes. Using MIMIC-IV data and a conformal prediction framework to limit uncertainty, a deep learning model specifically designed for non-ICU situations was created in this work. By enabling precise, early sepsis prediction with little data, the suggested methodology significantly improves resource allocation in hospital settings. A varied model distribution, inconsistent parameter use, and disparate quality assessments were noted.This thorough analysis emphasizes the significance of DL techniques for sepsis detection and early prediction utilizing EHR data.
References
1. Nemati, Shamim, Andre Holder, Fereshteh Razmi, Matthew D. Stanley, Gari D. Clifford, and Timothy G. Buchman. "An interpretable machine learning model for accurate prediction of sepsis in the ICU." Critical care medicine 46, no. 4 (2018): 547-553.
2. Wang, Dong, Jinbo Li, Yali Sun, Xianfei Ding, Xiaojuan Zhang, Shaohua Liu, Bing Han, Haixu Wang, Xiaoguang Duan, and Tongwen Sun. "A machine learning model for accurate prediction of sepsis in ICU patients." Frontiers in public health 9 (2021): 754348.
3. McCoy, Andrea, and Ritankar Das. "Reducing patient mortality, length of stay and readmissions through machine learning-based sepsis prediction in the emergency department, intensive care unit and hospital floor units." BMJ open quality 6, no. 2 (2017): e000158.
4. Persson, Inger, Andreas Östling, Martin Arlbrandt, Joakim Söderberg, and David Becedas. "A machine learning sepsis prediction algorithm for intended intensive care unit use (NAVOY Sepsis): proof-of-concept study." JMIR formative research 5, no. 9 (2021): e28000.
5. de Araújo, André Luiz Vale, Flavio Secco Fonseca, Arianne Sarmento Torcate, Ana Clara Gomes da Silva, Flavio Monteiro de Oliveira Júnior, José Carlos da Silva Júnior, Dayane Aparecida Gomes et al. "Early prediction of sepsis in intensive care units: A comparative analysis based on optimization techniques and committees." Research on Biomedical Engineering 41, no. 4 (2025): 55.
6. El-Rashidy, Nora, Tamer Abuhmed, Louai Alarabi, Hazem M. El-Bakry, Samir Abdelrazek, Farman Ali, and Shaker El-Sappagh. "Sepsis prediction in intensive care unit based on genetic feature optimization and stacked deep ensemble learning." Neural Computing and Applications 34, no. 5 (2022): 3603-3632.
7. Yuan, Kuo-Ching, Lung-Wen Tsai, Ko-Han Lee, Yi-Wei Cheng, Shou-Chieh Hsu, Yu-Sheng Lo, and Ray-Jade Chen. "The development an artificial intelligence algorithm for early sepsis diagnosis in the intensive care unit." International journal of medical informatics 141 (2020): 104176.
8. Lukaszewski, Roman A., Adam M. Yates, Matthew C. Jackson, Kevin Swingler, John M. Scherer, A. J. Simpson, Paul Sadler et al. "Presymptomatic prediction of sepsis in intensive care unit patients." Clinical and Vaccine Immunology 15, no. 7 (2008): 1089-1094.
9. Su, Longxiang, Zheng Xu, Fengxiang Chang, Yingying Ma, Shengjun Liu, Huizhen Jiang, Hao Wang et al. "Early prediction of mortality, severity, and length of stay in the intensive care unit of sepsis patients based on sepsis 3.0 by machine learning models." Frontiers in Medicine 8 (2021): 664966.
10. Moor, Michael, Bastian Rieck, Max Horn, Catherine R. Jutzeler, and Karsten Borgwardt. "Early prediction of sepsis in the ICU using machine learning: a systematic review." Frontiers in medicine 8 (2021): 607952.
11. Zhao, Rui, Tao Wan, Deyu Li, Zhengbo Zhang, and Zengchang Qin. "A deep learning model for early prediction of sepsis from intensive care unit records." In International Conference on Neural Information Processing, pp. 791-798. Cham: Springer International Publishing, 2020.
12. Lauritsen, Simon Meyer, Mads Ellersgaard Kalør, Emil Lund Kongsgaard, Katrine Meyer Lauritsen, Marianne Johansson Jørgensen, Jeppe Lange, and Bo Thiesson. "Early detection of sepsis utilizing deep learning on electronic health record event sequences." Artificial Intelligence in Medicine 104 (2020): 101820.
13. Al-Mualemi, Bilal Yaseen, and Lu Lu. "A deep learning-based sepsis estimation scheme." Ieee Access 9 (2020): 5442-5452.
14. Moor, Michael, Bastian Rieck, Max Horn, Catherine R. Jutzeler, and Karsten Borgwardt. "Early prediction of sepsis in the ICU using machine learning: a systematic review." Frontiers in medicine 8 (2021): 607952.
15. Awad, Aya, Mohamed Bader-El-Den, James McNicholas, Jim Briggs, and Yasser El-Sonbaty. "Predicting hospital mortality for intensive care unit patients: time-series analysis." Health informatics journal 26, no. 2 (2020): 1043-1059.
16. Desautels, Thomas, Jacob Calvert, Jana Hoffman, Melissa Jay, Yaniv Kerem, Lisa Shieh, David Shimabukuro et al. "Prediction of sepsis in the intensive care unit with minimal electronic health record data: a machine learning approach." JMIR medical informatics 4, no. 3 (2016): e5909.
17. Kim, Taehwa, Yunwon Tae, Hye Ju Yeo, Jin Ho Jang, Kyungjae Cho, Dongjoon Yoo, Yeha Lee et al. "Development and validation of deep-learning-based sepsis and septic shock early prediction system (DeepSEPS) using real-world ICU data." Journal of Clinical Medicine 12, no. 22 (2023): 7156.
18. McCoy, Andrea, and Ritankar Das. "Reducing patient mortality, length of stay and readmissions through machine learning-based sepsis prediction in the emergency department, intensive care unit and hospital floor units." BMJ open quality 6, no. 2 (2017): e000158.
19. Zhang, Yang, Juanjuan Hu, Tianfeng Hua, Jin Zhang, Zhongheng Zhang, and Min Yang. "Development of a machine learning-based prediction model for sepsis-associated delirium in the intensive care unit." Scientific reports 13, no. 1 (2023): 12697.
20. Deng, Yuhan, Shuang Liu, Ziyao Wang, Yuxin Wang, Yong Jiang, and Baohua Liu. "Explainable time-series deep learning models for the prediction of mortality, prolonged length of stay and 30-day readmission in intensive care patients." Frontiers in Medicine 9 (2022): 933037.
21. Li, Xiang, Xiao Xu, Fei Xie, Xian Xu, Yuyao Sun, Xiaoshuang Liu, Xiaoyu Jia et al. "A time-phased machine learning model for real-time prediction of sepsis in critical care." Critical care medicine 48, no. 10 (2020): e884-e888.
