Uncertainty-Aware Temporal Transformers for Early Prediction of Patient Deterioration from Irregular Clinical Time Series
Article Sidebar
Main Article Content
Abstract
Early identification of patient deterioration is essential for timely clinical intervention, yet prediction from electronic health records is complicated by irregular sampling, missing observations, and substantial uncertainty in longitudinal measurements. This study proposes an uncertainty-aware temporal Transformer for early prediction of clinical deterioration from multivariate medical time-series data. The framework introduces time-aware embeddings to represent irregular intervals between observations and employs hierarchical temporal attention to capture both short-term physiological changes and long-term patient trajectories. An uncertainty estimation module jointly models predictive confidence and dynamically reduces the influence of noisy or sparsely observed variables. To improve robustness, the model is trained with temporal masking and perturbation-based augmentation that simulate realistic missing-data patterns. Experiments on clinical time-series datasets demonstrate improved discrimination, calibration, and early-warning performance compared with recurrent, Transformer-based, and conventional risk prediction models. These findings highlight the potential of uncertainty-aware temporal modeling for reliable early-warning systems in data-intensive clinical environments.
Article Details
This work is licensed under a Creative Commons Attribution 4.0 International License.
Mind forge Academia also operates under the Creative Commons Licence CC-BY 4.0. This allows for copy and redistribute the material in any medium or format for any purpose, even commercially. The premise is that you must provide appropriate citation information.
References
X. Yan, Y. Jiang, W. Liu, D. Yi, and J. Wei, “Transforming Multidimensional Time Series into Interpretable Event Sequences for Advanced Data Mining,” ICHCI, pp. 126–130, 2024.
B. Lim, S. Ö. Arık, N. Loeff, and T. Pfister, “Temporal Fusion Transformers for interpretable multi-horizon time series forecasting,” International Journal of Forecasting, vol. 37, no. 4, pp. 1748–1764, 2021.
X. Yan, W. Wang, M. Xiao, Y. Li, and M. Gao, “Survival prediction across diverse cancer types using neural networks,” pp. 134–138, 2024.
S. N. Shukla and B. M. Marlin, “Multi-Time Attention Networks for Irregularly Sampled Time Series,” ICLR, 2021.
H. Zheng, Y. Ma, Y. Wang, G. Liu, Z. Qi, and X. Yan, “Structuring low-rank adaptation with semantic guidance for model fine-tuning,” ICECAI, pp. 731–735, 2025.
R. Miotto, L. Li, B. A. Kidd, and J. T. Dudley, “Deep Patient: An unsupervised representation to predict the future of patients from the electronic health records,” Scientific Reports, vol. 6, Art. no. 26094, 2016.
J. Wei, Y. Liu, X. Huang, X. Zhang, W. Liu, and X. Yan, “Self-Supervised Graph Neural Networks for Enhanced Feature Extraction in Heterogeneous Information Networks,” ICMLCA, pp. 272–276, 2024.
H. Harutyunyan et al., “Multitask learning and benchmarking with clinical time series data,” Scientific Data, vol. 6, Art. no. 96, 2019.
W. Wang, Y. Li, X. Yan, M. Xiao, and M. Gao, “Breast cancer image classification method based on deep transfer learning,” pp. 190–197, 2024.
H. Zheng, L. Zhu, W. Cui, R. Pan, X. Yan, and Y. Xing, “Selective knowledge injection via adapter modules in large-scale language models,” ICAIDE, pp. 373–377, 2025.
A. Rajkomar et al., “Scalable and accurate deep learning with electronic health records,” npj Digital Medicine, vol. 1, Art. no. 18, 2018.
Y. Li, W. Zhao, B. Dang, X. Yan, M. Gao, W. Wang, and M. Xiao, “Research on adverse drug reaction prediction model combining knowledge graph embedding and deep learning,” MLISE, pp. 322–329, 2024.
X. Yan, J. Du, L. Wang, Y. Liang, J. Hu, and B. Wang, “The Synergistic Role of Deep Learning and Neural Architecture Search in Advancing Artificial Intelligence,” ICEDCS, pp. 452–456, 2024.
B. Shickel et al., “Deep EHR: A survey of recent advances in deep learning techniques for electronic health record analysis,” IEEE Journal of Biomedical and Health Informatics, vol. 22, no. 5, pp. 1589–1604, 2018.
X. Yan, J. Du, X. Li, X. Wang, X. Sun, P. Li, and H. Zheng, “A Hierarchical Feature Fusion and Dynamic Collaboration Framework for Robust Small Target Detection,” IEEE Access, vol. 13, pp. 92953–92964, 2025.
E. Choi et al., “RETAIN: An interpretable predictive model for healthcare using reverse time attention mechanism,” NeurIPS, 2016.
M. Xiao, Y. Li, X. Yan, M. Gao, and W. Wang, “Convolutional neural network classification of cancer cytopathology images: Taking breast cancer as an example,” pp. 145–149, 2024.
Y. Li, X. Yan, M. Xiao, W. Wang, and F. Zhang, “Investigation of Creating Accessibility Linked Data Based on Publicly Available Accessibility Datasets,” pp. 77–81, 2024.