相似的预测拟合但不同的潜在动力学：刻画脑疾病个性化模型中学习到的动力学结构
Similar Predictive Fit but Different Latent Dynamics: Characterizing Learned Dynamical Structure in Personalized Models of Brain Disorders
浏览论文内容
中文总结 AI 辅助
该研究针对脑疾病个性化模型，发现即使预测拟合相似，癫痫与非癫痫组的学习动力学结构仍存在显著差异，表明需结合学习结构评估个性化临床模型。
中文摘要 AI 辅助
随着AI模型向临床决策和个性化治疗方向发展，理解模型学习到的内容除了预测准确性之外也十分重要。本研究探究，即使预测拟合相似，个性化潜在动力学是否仍能揭示与临床相关的差异。研究使用在Temple大学脑电图语料库（TUEG）上预训练的轻量型CNN-Transformer脑电图基础模型，提取片段级表征；利用Temple大学癫痫语料库（TUEP），将表征映射到共享潜在状态空间，并为每个受试者独立拟合稀疏多项逻辑转换分布（mLTD），以获得个性化转换依赖图Wₙ。分析涵盖n=198名受试者（99名癫痫患者、99名非癫痫患者）。在k=4时，癫痫受试者表现出显著更密集的学习依赖结构（p=1.1×10⁻⁷）；k=6时也呈现相同模式（19.90 vs.13.46；p=5.2×10⁻⁵）。基于图的特征在5折受试者交叉验证下提供中等程度的组间区分度（k=4时AUROC为0.68；k=6时为0.65）。相比之下，k=4时两组的保留对数似然几乎相同（-0.992 vs.-0.991；p=0.95），下一个状态预测也匹配良好（AUROC为0.855 vs.0.861；p=0.54）。因此，相似的预测拟合并不意味着相似的学习动力学：各组可具有相当的可预测性，同时个性化模型学习到的内部动力学结构存在显著差异。这一区别促使在个性化临床模型中，需结合预测性能对学习到的结构进行评估。
英文摘要
As AI models move toward clinical decision-making and personalized treatment, understanding \emph{what} a model learns is important beyond predictive accuracy alone. We investigate whether personalized latent dynamics reveal clinically associated differences even when predictive fit is similar. A lightweight CNN--Transformer EEG foundation model pretrained on the Temple University EEG Corpus (TUEG) extracts segment-level representations. Using the Temple University Epilepsy Corpus (TUEP), representations are mapped to a shared latent-state space, and sparse multinomial logistic transition distributions (mLTD) are fit independently to each subject to obtain personalized transition-dependency graphs $W_n$. Analyses include $n{=}198$ subjects (99 epilepsy / 99 non-epilepsy). At $k{=}4$, epilepsy subjects exhibit substantially denser learned dependency structure ($p{=}1.1\times10^{-7}$), with the same pattern at $k{=}6$ (19.90 vs. 13.46; $p{=}5.2\times10^{-5}$). Graph-derived features provide moderate group discrimination under 5-fold subject-wise cross-validation (AUROC 0.68 at $k{=}4$; 0.65 at $k{=}6$). In contrast, held-out log-likelihood is nearly identical between groups at $k{=}4$ ($-0.992$ vs. $-0.991$; $p{=}0.95$), with similarly matched next-state prediction (AUROC 0.855 vs. 0.861; $p{=}0.54$). Thus, similar predictive fit does not imply similar learned dynamics: groups can be comparably predictable while differing substantially in the internal dynamical structure learned by personalized models. This distinction motivates evaluating learned structure alongside predictive performance in personalized clinical models.
发表机构
- University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）
- Carle Foundation Hospital（卡尔基金会医院）
机构由 AI 辅助整理，请以论文原文为准。
