Robust Temporal Pattern Mining for Early Detection of Acute Clinical Events in ICU Settings Using LSTM Variants
DOI:
https://doi.org/10.56294/mw2025898Keywords:
Anomaly detection, Attention mechanisms, Deep temporal modeling, ICU monitoring, Predictive analytics, Robust time-series learning, Temporal embeddings, Temporal pattern mining, Calibration analysis, Clinical event predictionAbstract
Introduction: Anomaly detection in the intensive care unit (ICU) is very important for the detection of acute events in the initial stages. However, the sensitivity of the current methods based on time series for anomaly detection in the ICU is low.
Objective: To build an exhaustive anomaly-conscious learning system for the temporal pattern discovery process in the ICU environment to better identify critical incidents within the ICU earlier in time while being highly robust, interpretable, and computationally efficient.
Method: The proposed architecture combines convolutional, attention-based concepts, and recurrence for temporal representation in an integrated approach. It uses attention-based models for the representation of sequences in the ICU data, multi-scale representations for temporal embeddings, and refined representations for anomalous areas in the input ICU data. Such an approach achieves continuous representation of time-series physiology for the detection of anomalous regions.
Results: The proposed framework emerges consistently better than the state-of-the-art solutions such as LSTM networks, GRU-D networks, TCN networks, Informer networks, TFT networks, TimesNet networks, and PatchTST networks on all prominent performance measures. It presents an F1-score of 0.924, an AUPRC of 0.941, an ECE of 1.8%, and an overall accuracy of 93.4%. Even in the context of the domain shift problem, the degradation in performance is minimized with the ΔAUC of 1.5% and ΔECE of 0.4%. However, the proposed framework consumes just 0.9 joules per inference through the processing of 38 sequences per second in 92 milliseconds. Weighted scoring for anomalous cases also helps significantly in their immediate detection.
Conclusion: Through the careful integration of the goals of accuracy, understandability, and computational efficiency, the proposed anomaly-aware model achieves the state-of-the-art performance for the task of early warning in critical care. Its high-level predictive ability, robustness against the problem of domain shift, along with low computational complexity, make it very promising for its real-time implementation within the ICU setting.
References
[1] Sharafoddini A, Dubin JA, Lee J. Identifying subpopulations of septic patients: A temporal data-driven approach. Comput Biol Med. 2021;130:104182. DOI: https://doi.org/10.1016/j.compbiomed.2020.104182
[2] Chen W, Long G, Yao L, Sheng QZ. AMRNN: Attended multi-task recurrent neural networks for dynamic illness severity prediction. World Wide Web. 2020;23:2753–2770. DOI: https://doi.org/10.1007/s11280-019-00720-x
[3] Aldosari B. Supportive care pathway functionalities of EHR system in a Saudi Arabian hospital. Comput Biol Med. 2017;89:190–196. DOI: https://doi.org/10.1016/j.compbiomed.2017.08.012
[4] Magnin B, Mesrob L, Kinkingnéhun S, Pélégrini-Issac M, Colliot O, Sarazin M, et al. Support vector machine-based classification of Alzheimer’s disease from whole-brain anatomical MRI. Neuroradiology. 2009;51:73–83. DOI: https://doi.org/10.1007/s00234-008-0463-x
[5] Asadi H, Dowling R, Mitchell P. Machine learning for outcome prediction of acute ischemic stroke post intra-arterial therapy. PLoS One. 2014;9:e88225. DOI: https://doi.org/10.1371/journal.pone.0088225
[6] Malhotra C, Dilley E, Han-Burgess JN, Valdez J, Robertson J, Clark C, et al. Predicting drug-resistant epilepsy—A machine learning approach based on administrative claims data. Epilepsy Behav. 2018;89:118–125. DOI: https://doi.org/10.1016/j.yebeh.2018.10.013
[7] Vincent JL, Moreno R, Takala J, Willatts S, De Mendonça A, Bruining H, et al. The SOFA (Sepsis-Related Organ Failure Assessment) Score to describe organ dysfunction/failure. Berlin/Heidelberg: Springer; 1996. DOI: https://doi.org/10.1007/s001340050156
[8] Mardini L, Lipes J, Jayaraman D. Adverse outcomes associated with delayed intensive care consultation in medical and surgical inpatients. J Crit Care. 2012;27:688–693. DOI: https://doi.org/10.1016/j.jcrc.2012.04.011
[9] Schein RM, Hazday N, Pena M, Ruben BH, Sprung CL. Clinical antecedents to in-hospital cardiopulmonary arrest. Chest. 1990;98:1388–1392. DOI: https://doi.org/10.1378/chest.98.6.1388
[10] Knaus WA, Draper EA, Wagner DP, Zimmerman JE. APACHE II: A severity of disease classification system. Crit Care Med. 1985;13:818–829. DOI: https://doi.org/10.1097/00003246-198510000-00009
[11] Knaus WA, Wagner DP, Draper EA, Zimmerman JE, Bergner M, Bastos PG, et al. The APACHE III prognostic system: Risk prediction of hospital mortality for critically ill hospitalized adults. Chest. 1991;100:1619–1636. DOI: https://doi.org/10.1378/chest.100.6.1619
[12] Le Gall JR, Lemeshow S, Saulnier F. A new simplified acute physiology score (SAPS II) based on a European/North American multicenter study. JAMA. 1993;270:2957–2963. DOI: https://doi.org/10.1001/jama.1993.03510240069035
[13] Moreno RP, Metnitz PGH, Almeida E, Jordan B, Bauer P, Campos RA, et al. SAPS 3—From evaluation of the patient to evaluation of the intensive care unit. Part 2: Development of a prognostic model for hospital mortality at ICU admission. Intensive Care Med. 2005;31:1345–1355. DOI: https://doi.org/10.1007/s00134-005-2763-5
[14] Kotwal JG, Kashyap R, Shafi PM. Yolov5-based convolutional feature attention neural network for plant disease classification. Int J Intell Syst Technol Appl. 2024;22(3):237–259. doi:10.1504/IJISTA.2024.140949 DOI: https://doi.org/10.1504/IJISTA.2024.140949
[15] Diwan P, Khandelwal B. Blockchain assisted encryption scheme for intellectual share estimation using medical research data. Concurrency Comput Pract Exp. 2024;36(2). doi:10.1002/cpe.7896 DOI: https://doi.org/10.1002/cpe.7896
[16] Kashyap R. Advanced diagnosis techniques in medical imaging. In: Challenges and Applications for Implementing Machine Learning in Computer Vision. 2019. p. 64–91. doi:10.4018/978-1-7998-0182-5.ch003 DOI: https://doi.org/10.4018/978-1-7998-0182-5.ch003
[17] Shah SY, Yuan Z, Lu S, Zerfos P. Dependency analysis of cloud applications for performance monitoring using recurrent neural networks. In: Proc IEEE Int Conf Big Data. Boston (MA): IEEE; 2017. p. 1534–1543. DOI: https://doi.org/10.1109/BigData.2017.8258087
[18] Hewamalage H, Bergmeir C, Bandara K. Recurrent neural networks for time series forecasting: Current status and future directions. Int J Forecast. 2021;37:388–427. DOI: https://doi.org/10.1016/j.ijforecast.2020.06.008
[19] Ahmad S, Neal Joshua ES, Rao NT, Ghoniem RM, Taye BM, Bharany S. A multi stage deep learning model for accurate segmentation and classification of breast lesions in mammography. Scientific Reports. 2025 Oct 23;15(1):37103 DOI: https://doi.org/10.1038/s41598-025-21146-8
[20] Nair R, Abdulhasan MM, Khalaf HH, Shareef AM. A deep learning-based model for mutation rate prediction of COVID-19 using genomic sequences. In: 7th Int Conf Image Information Processing (ICIIP). Solan, India: IEEE; 2023. p. 759–764. doi:10.1109/ICIIP61524.2023.10537657 DOI: https://doi.org/10.1109/ICIIP61524.2023.10537657
[21] Ionescu CM, Keyser RD, Copot D, Yumuk E, Ynineb A, Othman GB, Neckebroek M. Model extraction from clinical data subject to large uncertainties and poor identifiability. IEEE Control Syst Lett. 2020;1.
[22] Pham TD. Time–frequency time–space LSTM for robust classification of physiological signals. Sci Rep. 2021;11:6936. DOI: https://doi.org/10.1038/s41598-021-86432-7
[23] Heynen J, Copot D, Ghita M, Ionescu CM. Using convolutional neural network online estimators for predicting pain-level variability enables predictive control of anesthesia. In: Proceedings of the 25th International Conference on System Theory, Control and Computing (ICSTCC); 2021 Oct 20–23; Iași, Romania. p. 194–199. DOI: https://doi.org/10.1109/ICSTCC52150.2021.9607270
[24] Gundu V, Simon SP. PSO–LSTM for short term forecast of heterogeneous time series electricity price signals. J Ambient Intell Humaniz Comput. 2021;12:2375–85. DOI: https://doi.org/10.1007/s12652-020-02353-9
[25] Khorram A, Khalooei M, Rezghi M. End-to-end CNN+LSTM deep learning approach for bearing fault diagnosis. Appl Intell. 2021;51:736–51. DOI: https://doi.org/10.1007/s10489-020-01859-1
[26] Almadhor A, Ojo S, Nathaniel TI, Ahmad S, Hejaili AA. “Deep feature-driven SVM model with XAI for reliable colorectal cancer imaging analysis”, Signal, Image and Video Processing. 2025 Dec;19(15):1-8. DOI: https://doi.org/10.1007/s11760-025-04885-z
[27] Rajawat AS, Ahmad S, Muqeem M, Abdeljaber HA, Alanazi S, Nazeer J. “Advanced Deep Learning Integration for Early Pneumonia Detection for Smart Healthcare”, International Journal of Online & Biomedical Engineering. 2025 Mar 1;21(3). DOI: https://doi.org/10.3991/ijoe.v21i03.53107
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Ahmed A.F Osman, Sultan Ahmad, Rajit Nair, Ramgopal Kashyap, Mosleh Hmoud Al-Adhaileh, Theyazn H.H Aldhyani, Hikmat A. M. Abdeljaber, Maryam Nasser Almusallam (Author)
This work is licensed under a Creative Commons Attribution 4.0 International License.
The article is distributed under the Creative Commons Attribution 4.0 License. Unless otherwise stated, associated published material is distributed under the same licence.
