Data Quality-Aware Deep Learning for Reliable Clinical Prediction from Noisy Electronic Health Records
- Authors
-
-
Soren Valberg
California State University
Author
-
- Abstract
-
Electronic health records (EHRs) provide large-scale data for clinical prediction, but their practical utility is often limited by missing values, inconsistent measurements, coding errors, and noisy observations. This study proposes a data quality-aware deep learning framework for reliable clinical prediction from imperfect EHR data. Instead of treating all observations equally, the framework constructs patient-level and feature-level quality indicators based on missingness, temporal consistency, measurement frequency, and abnormal-value patterns. These indicators are jointly encoded with longitudinal clinical variables through a quality-guided attention network that dynamically reduces the contribution of unreliable observations. A self-supervised reconstruction objective is further introduced to learn robust representations from partially observed patient trajectories. Experiments are conducted on large-scale intensive care EHR data for mortality and clinical deterioration prediction under different levels of synthetic and naturally occurring data corruption. Results show that incorporating explicit data-quality information improves both predictive discrimination and calibration, particularly when 20–40% of clinical observations are missing or corrupted. The proposed approach demonstrates that modeling data reliability directly can improve the robustness of AI systems operating on real-world clinical records.
- References
-
[1] B. Shickel, P. J. Tighe, A. Bihorac, and P. Rashidi, “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.
[2] H. Zheng, Y. Ma, Y. Wang, G. Liu, Z. Qi, and X. Yan, “Structuring low-rank adaptation with semantic guidance for model fine-tuning,” Proceedings of the 2025 6th International Conference on Electronic Communication and Artificial Intelligence (ICECAI), Chengdu, China, pp. 731–735, 2025.
[3] X. Yan, W. Wang, M. Xiao, Y. Li, and M. Gao, “Survival prediction across diverse cancer types using neural networks,” Proceedings of the 2024 7th International Conference on Machine Vision and Applications, pp. 134–138, 2024.
[4] Z. C. Lipton, D. C. Kale, and R. Wetzel, “Modeling missing data in clinical time series with RNNs,” Machine Learning for Healthcare, 2016.
[5] 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,” 2024 5th International Conference on Machine Learning and Computer Application (ICMLCA), pp. 272–276, 2024.
[6] A. Rajkomar et al., “Scalable and accurate deep learning with electronic health records,” npj Digital Medicine, vol. 1, Art. no. 18, 2018.
[7] 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,” Proceedings of the 2024 7th International Conference on Machine Vision and Applications, pp. 145–149, 2024.
[8] 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.
[9] H. Harutyunyan, H. Khachatrian, D. C. Kale, G. Ver Steeg, and A. Galstyan, “Multitask learning and benchmarking with clinical time series data,” Scientific Data, vol. 6, Art. no. 96, 2019.
[10] 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.
[11] 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.
[12] 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.
[13] 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.
[14] W. Wang, Y. Li, X. Yan, M. Xiao, and M. Gao, “Breast cancer image classification method based on deep transfer learning,” International Conference on Image Processing, Machine Learning and Pattern Recognition, pp. 190–197, 2024.
[15] Y. Li, X. Yan, M. Xiao, W. Wang, and F. Zhang, “Investigation of Creating Accessibility Linked Data Based on Publicly Available Accessibility Datasets,” Proceedings of the 2023 13th International Conference on Communication and Network Security, pp. 77–81, 2024.
[16] 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.
[17] E. Choi et al., “RETAIN: An interpretable predictive model for healthcare using reverse time attention mechanism,” Advances in Neural Information Processing Systems, 2016.
[18] J. N. Acosta, G. J. Falcone, P. Rajpurkar, and E. J. Topol, “Multimodal biomedical AI,” Nature Medicine, vol. 28, pp. 1773–1784, 2022.
- Downloads
- Published
- 2026-02-01
- Issue
- Vol. 6 No. 2 (2026)
- Section
- Articles