针对亚组的患者适应性偏移以实现可靠诊断
Patient-Conditioned Adaptive Offsets for Reliable Diagnosis across Subgroups
- Department of Computer Science and Engineering, University of Notre Dame（计算机科学与工程系，诺丁汉大学）
- Department of Electrical Engineering, University of Notre Dame（电气工程系，诺丁汉大学）
- Weill Cornell Medicine（韦尔医学院）
- Department of Computer Science, Emory University（计算机科学系，埃默里大学）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
本文提出HyperAdapt框架，通过患者条件适应提升亚组诊断可靠性，同时保持共享模型的准确性。
AI中文摘要：
人工智能模型在医疗诊断中往往由于疾病发病率、影像表现和临床风险特征的异质性而在患者群体间表现出不均衡的性能。现有的算法公平性方法通常试图通过抑制敏感属性来减少这种差异。然而，在医疗环境中这些属性往往包含必要的诊断信息，去除它们会降低准确性和可靠性，特别是在高风险应用中。相反，临床决策在解释诊断证据时会明确纳入患者背景，这表明了对亚组意识模型的不同设计方向。在本文中，我们介绍了HyperAdapt，一种患者条件适应框架，它在保持共享诊断模型的同时提高亚组的可靠性。临床相关的属性如年龄和性别被编码成紧凑的嵌入，并用于条件化一个超网络式模块，该模块为共享骨干网络的选定层生成小残差调节参数。这种设计保留了骨干网络学习的一般医学知识，同时使能够针对患者特定的变异性进行针对性调整。为了确保效率和鲁棒性，适应性通过低秩和瓶颈参数化进行限制，从而限制了模型复杂性和计算开销。在多个公共医学影像基准测试中，实验表明所提出的方法在不牺牲整体准确性的情况下一致提高了亚组水平的性能。在PAD-UFES-20数据集上，我们的方法在召回率上比最强的竞争对手基线高出4.1%，在F1分数上高出4.4%，对于少数群体患者显示出更大的收益。
英文摘要：
AI models for medical diagnosis often exhibit uneven performance across patient populations due to heterogeneity in disease prevalence, imaging appearance, and clinical risk profiles. Existing algorithmic fairness approaches typically seek to reduce such disparities by suppressing sensitive attributes. However, in medical settings these attributes often carry essential diagnostic information, and removing them can degrade accuracy and reliability, particularly in high-stakes applications. In contrast, clinical decision making explicitly incorporates patient context when interpreting diagnostic evidence, suggesting a different design direction for subgroup-aware models. In this paper, we introduce HyperAdapt, a patient-conditioned adaptation framework that improves subgroup reliability while maintaining a shared diagnostic model. Clinically relevant attributes such as age and sex are encoded into a compact embedding and used to condition a hypernetwork-style module, which generates small residual modulation parameters for selected layers of a shared backbone. This design preserves the general medical knowledge learned by the backbone while enabling targeted adjustments that reflect patient-specific variability. To ensure efficiency and robustness, adaptations are constrained through low-rank and bottlenecked parameterizations, limiting both model complexity and computational overhead. Experiments across multiple public medical imaging benchmarks demonstrate that the proposed approach consistently improves subgroup-level performance without sacrificing overall accuracy. On the PAD-UFES-20 dataset, our method outperforms the strongest competing baseline by 4.1% in recall and 4.4% in F1 score, with larger gains observed for underrepresented patient populations.
