跳到正文
返回论文列表
stat.ME提交于 已译

风险感知分位数学习用于个性化动态治疗方案

Risk-Aware Quantile Learning for Personalized Dynamic Treatment Regimes

Chunyin Lei · Annie Qu

中文摘要

序贯临床决策往往不只是最大化平均疗效。临床医生可能需要同时优化结局分布中临床相关的尾部、控制治疗相关风险,并在多个治疗选项间选择。现有的分位数动态治疗方案(DTR)方法能刻画治疗结局的分布特征,但大多局限于仅关注疗效的目标和二分类治疗。为此,提出风险感知分位数动态治疗方案(RQDTR),一个统一框架,在优化累积潜在结局的预设分位数的同时显式纳入治疗相关风险。还开发了一种基于角度的表述,用于跨多个治疗类别联合学习决策规则。该框架包含三个可解释的子类:仅疗效的分位数学习、带群体水平风险控制的约束式学习,以及通过复合获益-风险效用进行的效用式学习。理论上,建立了可识别性与 oracle 等价性、在递增二次惩罚下的替代与估计量一致性、渐近风险可行性,以及有限样本性能误差界。大量模拟研究以及在 All of Us 重性抑郁障碍和 MIMIC-III 脓毒症数据上的应用表明,RQDTR 相比现有分位数 DTR 方法改善了面向尾部的疗效,并实现了更有利的获益-风险权衡。

关键要点

  1. 01现有分位数 DTR 方法多局限于仅疗效目标和二分类治疗,无法同时控制治疗相关风险。
  2. 02提出 RQDTR 统一框架,优化累积潜在结局的预设分位数并显式纳入治疗相关风险。
  3. 03给出基于角度的表述,支持跨多个治疗类别联合学习决策规则。
  4. 04建立可识别性、oracle 等价性、估计量一致性、渐近风险可行性与有限样本误差界。
  5. 05在 All of Us 抑郁与 MIMIC-III 脓毒症数据上改善尾部疗效并取得更优获益-风险权衡。

解读

尚无解读。

原始英文摘要

arXiv:2608.05434v2 Announce Type: replace-cross Abstract: Sequential clinical decision-making often involves more than maximizing average efficacy. Clinicians may need to simultaneously optimize clinically relevant tails of the outcome distribution, control treatment-related risk, and choose among multiple treatment options. Existing quantile dynamic treatment regime (DTR) methods capture distributional features of treatment outcomes but remain largely restricted to efficacy-only objectives and binary treatments. To address these limitations, we propose Risk-Aware Quantile Dynamic Treatment Regimes (RQDTR), a unified framework that optimizes a prespecified quantile of the cumulative potential outcome while explicitly incorporating treatment-related risk. We also develop an angle-based formulation for jointly learning decision rules across multiple treatment categories. Our framework includes three interpretable subclasses: efficacy-only quantile learning, constraint-based learning with population-level risk control, and utility-based learning through a composite benefit-risk utility. Theoretically, we establish identification and oracle equivalence, surrogate and estimator consistency under increasing quadratic penalties, asymptotic risk feasibility, and finite-sample performance error bounds. Extensive simulation studies and applications to All of Us major depressive disorder and MIMIC-III sepsis data demonstrate that RQDTR improves tail-oriented efficacy and achieves more favorable benefit-risk trade-offs than existing quantile DTR methods.

同方向论文 · stat.ME

查看全部 →