Saved in:
Bibliographic Details
Main Authors: Zhang, Peisong, Peng, Manqiang, Wu, Yuxuan, Phadungsaksawasdi, Pawit, Yeung, Wesley, Zhang, Ye, Nguyen, Trang, Zhang, Qiang, Liu, Nan, Wang, Meng, Ngiam, Kee Yuan, Tham, Yih-Chung, Cheng, Ching-Yu, Fu, Tianfan, Chen, Qingyu, Ke, Rosemary, Li, Chang, Yang, Wenzhuo, Lu, Zhenghao, Lai, Chunyou, Zhang, Yu, Zhong, Sheng, Deng, Hao, Liu, Dianbo
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2605.05706
Tags: Add Tag
No Tags, Be the first to tag this record!
Table of Contents:
  • Estimating individualized treatment effects from longitudinal observational data is central to data-driven medicine, yet existing methods face a fundamental limitation: reducing confounding bias often suppresses clinically informative heterogeneity, degrading patient-specific predictions. Here, we identify this tension as a bias-precision paradox in causal representation learning and introduce sampling-based maximum mean discrepancy (sMMD), a stochastic alignment strategy that replaces global adversarial balancing with subset-level matching. We instantiate this approach in a framework for counterfactual outcome prediction with attribution-grounded interpretability. Across two large-scale ICU cohorts (n = 27,783), our framework improves accuracy under distribution shift, reducing error by up to 11.5% and substantially increasing recall in high-risk tasks. Mechanistic analyses show that sMMD selectively preserves clinically decisive variables. In human-AI evaluation, our method outperforms clinicians-in-training and large language models, and improves clinician accuracy by 14.7% while reducing decision time, enabling interpretable, real-time clinical decision support.