IASCI Research Publishing
Translational Methods for Digital Healthcare
Research Article Published October 9, 2026 Article tmdh20260005

Causal Spurious Cue Removal for Robust Clinical Text Classification

Read & download PDF
Abstract

Clinical text classification models often depend on spurious textual cues, including department names, template phrases, physician writing habits, medication shortcuts, and institutional documentation patterns. Such shortcuts may inflate internal performance but reduce reliability when the model is applied to new hospitals or electronic health record systems. This study proposes a Causal Cue Removal Transformer (CCRT) for robust disease risk classification from clinical notes. The method first detects shortcut-prone tokens through environment-wise attention variation, counterfactual token replacement, and conditional mutual information estimation. It then reduces shortcut influence by applying causal token masking and invariant prediction regularization across hospitals. The model was evaluated on 214,600 de-identified clinical notes from 58,320 patients across four hospitals, including discharge summaries, emergency records, and outpatient notes. The task was 30-day readmission prediction and major complication classification. In internal testing, CCRT achieved an AUROC of 0.871, compared with 0.856 for ClinicalBERT. In cross-hospital validation, AUROC improved from 0.742 to 0.803, while calibration error decreased from 0.091 to 0.052. Error analysis showed a 36.4% reduction in predictions driven mainly by administrative or template-specific expressions. The findings show that causal cue removal can improve text model robustness in clinical deployment.

Keywords
Clinical text classificationshortcut learningcausal token maskingtransformer modelhospital shiftrobust prediction
References
  1. Chen, H., Ning, P., Li, J., & Mao, Y. (2025). Energy Consumption Analysis and Optimization of Speech Algorithms for Intelligent Terminals.
  2. Zahed, T. A., Mishra, B. K., & Fagbola, T. M. (2025, September). Domain-Specific Fine-Tuning of PubMedBERT for Neonatal Care Applications. In 2025 2nd International Conference on Circuits, Power and Intelligent Systems (CCPIS) (pp. 1-6). IEEE.
  3. Jiao, Y., Zhao, B., Wang, A., & Shi, T. (2026). Construction and Empirical Study of a Modularized Teaching System for Art Courses Based on a Unified Training Pathway.
  4. Zheng, J., & Makar, M. (2022). Causally motivated multi-shortcut identification and removal. Advances in Neural Information Processing Systems, 35, 12800-12812.
  5. Mini Han Wang , " AI-Powered Innovations in Ophthalmic Diagnosis and Treatment ", Bentham Science Publishers (2025). https://doi.org/10.2174/97988988122871250101.
  6. Zhang, Z. (2026). A Study on the Impact of Cost Presentation on Decision-making Bias and Cancellation Behavior in Online Subscriptions. Available at SSRN 7349839.
  7. Qi, C., & Qiao, X. (2026). Building Reliable AI Systems: A Framework That Bridges Architecture and Operations. Available at SSRN 6795298.
  8. Han, Z., Chen, W., Han, Y., Mao, R., & Qin, J. (2026). Fast Diversified Top-k Rule Discovery via User-Guided Embeddings. IEEE Transactions on Knowledge and Data Engineering.
  9. Li, Y., & Liu, S. (2026, May). A Study on Dynamic Optimization of Alerting Policies and Multi-Agent Decision-Making Mechanisms in Cloud Environments. In 2026 7th International Seminar on Artificial Intelligence, Networking and Information Technology (AINIT) (pp. 703-706). IEEE.
  10. Xiong, W., Zeng, Y., & Guo, Y. (2026). A Study on the Impact of MEP Space Organization in Large-Scale Commercial Complexes on Investment Efficiency and Its Mechanisms.
  11. Hong, Z., Liang, S., Xu, T., & Chen, H. (2026). A Study on Failover Verification and Recovery Objective Prediction for Cross-Region Cloud Services.
  12. Quan, Q., Gao, Y., & Wang, Q. (2025). The impact of teacher emotional support on students' engagement in AI-mediated English learning environments: The mediating role of resilience and self-efficacy. Acta Psychologica, 260, 105766.
  13. Huang, J., Xu, T., Yang, J., & Yin, J. (2026). Blockage Early Warning and Financial Impact Quantification in the Monthly Closing Process of Fintech Companies. Available at SSRN 7179198.
  14. Zhang, Z., Wang, J., Li, Z., Wang, Y., & Zheng, J. (2025). Anncoder: A mti-agent-based code generation and optimization model. Symmetry, 17(7), 1087.
  15. Zhao, J., Fan, J., & Li, L. (2026). A Study on an Explainable Causal-Enhanced LLM Agent for Predicting the Forming Quality of Automotive Component Materials.
  16. Wang, Z., Yu, J., Liu, H., Zheng, Z., Jin, Y., Li, S., ... & Zhang, K. (2025, July). Generative Music Models’ Alignment with Professional and Amateur Users’ Expectations. In Findings of the Association for Computational Linguistics: ACL 2025 (pp. 6909-6920).
  17. Jiao, Y., Shi, T., Zhao, B., & Wang, A. (2026). Retrieval-Guided Structured Reasoning and Interpretable Representation Learning for Large-Scale Video–Language Models.
  18. Liang, S., Du, Y., Chen, W., & Liu, Z. (2026). Order Allocation and Emergency Transportation Decisions Amid Multi-Source Procurement Disruptions.
  19. Bao, Y., Qiu, Y., & Wang, H. (2026). FLARE: A Real-Time Framework for Detecting Malicious Account Activities at Internet Scale. Available at SSRN 7185659.
  20. Xu, Y., Cui, J., Zhu, Z., Shang, H., Luan, S., Xu, M., ... & Zhu, S. (2026). Wam-flow: Parallel coarse-to-fine motion planning via discrete flow matching for autonomous driving. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 24918-24928).
  21. Wang, S., Feng, Y., & Fang, X. (2026, May). A Large Language Model-Enabled Multi-Agent Collaboration Method for Complex Task Solving. In 2026 6th International Symposium on Computer Technology and Information Science (ISCTIS) (pp. 253-256). IEEE.
  22. Yang, Q., & Du, Y. (2026). Predicting Rollback Risks and Controlling Gradual Rollout Traffic for Online Ranking Feature Deployments.
  23. Zheng, X., Chen, X., Gong, S., Griffin, X., & Slabaugh, G. (2025, September). Xfmamba: Cross-fusion mamba for multi-view medical image classification. In International Conference on Medical Image Computing and Computer-Assisted Intervention (pp. 672-682). Cham: Springer Nature Switzerland.
  24. Liang, S., Hsu, K. H., & Liu, Z. (2026). Rolling Scheduling and Capacity Balancing for Server Assembly Under Engineering Change Disturbances.
  25. Zhang, G., Xu, Y., Zhang, M., Wang, S., & Lin, N. (2021). The brain network in support of social semantic accumulation. Social cognitive and affective neuroscience, 16(4), 393-405.
  26. Su, J., He, C., Zhu, F., Xu, X., Guan, D., & Si, C. (2024, September). Hpff: Hierarchical locally supervised learning with patch feature fusion. In European Conference on Computer Vision (pp. 293-309). Cham: Springer Nature Switzerland.
  27. Yuan, Y., Huang, J., Yin, J., & Xu, T. (2026). Confidence Calibration and Semantic Error Analysis for Ambiguous Coreference Resolution in User-generated Social Media Text. Available at SSRN 7393238.
  28. Zhao, Z., & Welsch, R. E. (2024). Hierarchical reinforced trader (hrt): A bi-level approach for optimizing stock selection and execution. arXiv preprint arXiv:2410.14927.
  29. Lyu, Y. (2025). Enhancing Approximate Conformance Checking Accuracy with Hierarchical Clustering Model Behaviour Sampling. International Journal of Advanced Computer Science & Applications, 16(8).
  30. Yang, Z., Alexandrova, A. N., & Sautet, P. (2026). Modeling CO2 Hydrogenation to Methanol on an Ensemble of Inverse ZrO2 on Cu Catalytic Sites: Mechanism, Reactivity, and Deactivation. Angewandte Chemie, e5448247.
  31. Xu, Z., Zhang, X., Zhou, X., & Zhang, J. (2025). AvatarShield: Visual Reinforcement Learning for Human-Centric Video Forgery Detection. arXiv preprint arXiv:2505.15173.
  32. Xi, X., Chen, P., Qiu, Y., Luo, R., Tong, P., & Liang, J. Video-BCI: Bayesian Cognitive Integration of Self-Prior Hypotheses for Video Understanding. In Forty-third International Conference on Machine Learning.
  33. Liang, J., Xiao, Y., Yang, H., Yu, Z., Liu, J. N., & Yang, K. (2026, April). C-HyPOD: Causal Hyperbolic Representation Learning with Prototype Orthogonal Disentanglement for Graph Out-of-Distribution Recommendation. In Proceedings of the ACM Web Conference 2026 (pp. 6541-6550).
  34. Xu, Z., Zhang, X., Li, R., Tang, Z., Huang, Q., & Zhang, J. (2024). Fakeshield: Explainable image forgery detection and localization via multi-modal large language models. arXiv preprint arXiv:2410.02761.
  35. Huang, Q., Xu, Z., Zhang, X., & Zhang, J. (2025). UniShield: An Adaptive Multi-Agent Framework for Unified Forgery Image Detection and Localization. arXiv preprint arXiv:2510.03161.
  36. Jiang, C. (2026). Information doesn’t work? Testing experiential drivers of VR application in heritage tourism by affective-cognitive pathways. Journal of Hospitality and Tourism Technology, 1-28.
  37. Wang, T., & Xia, Z. (2025). Stability of In-Context Learning: A Spectral Coverage Perspective. arXiv preprint arXiv:2509.20677.
  38. Liu, T., He, J., Zhou, R., Song, J., Liu, H., & Liang, Y. (2025). Revolutionizing clinical decision making through deep learning and topic modeling for pathway optimization. Scientific Reports, 15(1), Article 28787. https://doi.org/10.1038/s41598-025-12679-z
  39. Wu, X., Xing, X., Yao, J., Qian, Q., & Song, J. (2025). Scnet: spectral convolutional networks for multivariate time series classification. Applied Intelligence, 55(6), Article 456. https://doi.org/10.1007/s10489-025-06352-1
  40. Ma, C., Song, J., Xu, Y., Fan, H., Wu, X., & Sun, T. (2024). Vehicle-Based Machine Vision Approaches in Intelligent Connected System. IEEE Transactions on Intelligent Transportation Systems, 25(3), 2827-2836. https://doi.org/10.1109/TITS.2023.3276325
  41. Chaoyong Ma, Nan Si, Xuewei Song, Chen Liang and Haoming Luo, A Bearing Fault Diagnosis Method Based on Concise Empirical Wavelet Transform, International Journal of Comprehensive Engineering，2023, 12(1), 1-6
Publication details
Journal
Translational Methods for Digital Healthcare
Volume
1 (2026)
Issue
1 · Forthcoming issue
Article number
tmdh20260005
License
CC BY 4.0