用逆强化学习提取帕金森病患者打字时的可解释运动偏好指标。
Inverse Reinforcement Learning for Interpretable Keystroke Biomarkers in Parkinson's Disease
- 通过逆强化学习从原始打字时间中恢复个体化运动速度偏好权重。
- 该权重与运动严重程度相关（r=-0.607，p<0.001），跨子组重复验证。
- 结果可解释且具高重测信度（ICC=0.903），适用于临床研究与远程监测。
打字动态提供了被动评估运动功能的窗口，但现有方法仅提取平均打字统计量并训练分类器进行帕金森病（PD）与健康对照区分，牺牲了可解释性，且极少报告可靠性。本文采用最大熵逆强化学习（IRL）处理原始打字时间，无需临床标签即可恢复每位受试者的速度偏好权重（w_speed），反映其对快速动作的隐含代价。在neuroQWERTY MIT-CSXPD数据集（85名受试者，42名PD）上，诊断并修正了初始四参数分解中的特征共线性问题，得到可识别的三参数模型。恢复的w_speed与UPDRS-III运动评分显著相关（r=-0.607，95% CI [-0.770, -0.364]，p<0.001，n=42），在两个独立子队列中复现（r=-0.720，r=-0.588），且在控制飞行时间均值与标准差后仍保持显著偏相关（r=-0.371，p=0.016）。该方法优于SHAP和LASSO在相同代理特征上的表现（r=+0.362，r=+0.410），并提供个体化可解释输出。模型无监督AUC为0.605，留一法交叉验证AUC达0.750（95% CI [0.644, 0.847]），证实其判别价值。跨诊室测试-重测信度极高（ICC(2,1)=0.903，95% CI [0.842, 0.971]），是首个报告正式可靠性的打字-PD研究。另外两个恢复权重（一致性、手交替）未通过混杂因素检验，增强了存活信号的可信度。在独立的mPower智能手机敲击数据（n=200）上实现跨模态外部验证（r=-0.639，p=2.52e-24，OR=14.19），确认了跨设备、跨国家的收敛效度。
原文摘要 · Abstract (English)
Keystroke dynamics offer a passive window into motor function, but existing work extracts aggregate typing statistics and trains classifiers for PD/control discrimination, foregoing interpretability and rarely reporting reliability. We instead apply maximum-entropy inverse reinforcement learning (IRL) to raw keystroke timing, recovering a per-subject speed-preference weight (w_speed) reflecting the implicit cost assigned to fast movement, without any clinical label during fitting. On the neuroQWERTY MIT-CSXPD dataset (85 subjects, 42 PD), we diagnose and correct a feature collinearity failure in an initial four-parameter decomposition, yielding an identifiable three-parameter model. The recovered w_speed correlates with UPDRS-III motor severity at r=-0.607 (95% CI [-0.770,-0.364], p<0.001, n=42), replicates across two independent sub-cohorts (r=-0.720, r=-0.588), and retains significant partial correlation after controlling for mean and SD of flight time (r=-0.371, p=0.016). It outperforms SHAP and LASSO on the same proxy features (r=+0.362 and r=+0.410) while additionally providing per-subject, interpretable output. A model-free AUC of 0.605 and LOO-CV AUC of 0.750 (95% CI [0.644,0.847]) confirm discriminative value. Test-retest reliability across clinic sessions yields ICC(2,1)=0.903 (95% CI [0.842,0.971]); no prior keystroke-PD study has reported formal reliability. Two other recovered weights (consistency, hand-alternation) did not survive confound checks, strengthening credibility of the surviving signal. Cross-modality external validation on independent mPower smartphone tapping data (n=200) recovers the same signal (r=-0.639, p=2.52e-24, OR=14.19), confirming convergent validity across modality, device, and country.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。