发表机构
Taipei Medical University; University of Memphis; Chitkara University; University College London; UPES; University of Eastern Finland; CNRS; University Bordeaux; Bordeaux INP（台北医学大学; 孟菲斯大学; 奇特卡拉大学; 伦敦大学学院; 石油与能源研究大学; 东芬兰大学; 法国国家科学研究中心; 波尔多大学; 波尔多国立理工学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究提出患者专属框架MCNet与MCS，量化AD中结构MRI与淀粉样PET的相对诊断贡献，在ADNI-3与OASIS-3队列中验证了其分期性能、单调梯度及跨队列泛化性，为痴呆护理可信AI奠定基础。
AI 中文摘要
结合结构MRI与正电子发射断层扫描（PET）的多模态神经成像可捕捉阿尔茨海默病（AD）连续谱中互补的结构-功能关系，但现有人工智能系统仅输出单一诊断标签，未针对特定患者量化哪种成像模态驱动了该决策。我们提出模态贡献网络（MCNet）与模态贡献分数（MCS），这是首个患者专属归因框架，可量化从认知正常到轻度认知障碍（MCI）再到AD连续谱中，模态主导性从结构萎缩向淀粉样与代谢功能障碍的转变。MCS通过模态消融对每个受试者归一化为1（对每个受试者i，MCS_MRI_i + MCS_PET_i = 1.0），提供了体液生物标志物无法提供的可解释、临床可操作的分数。该方法应用于327名ADNI-3参与者，其在认知正常、MCI、AD组中均衡分布，MCNet取得了有竞争力的三阶段分期性能（AUC=0.881）。MCS显示出统计学显著的单调梯度（Kruskal-Wallis检验p<0.0001），从认知正常（MCS_PET为0.412±0.229）到MCI（0.489±0.289）再到AD（0.671±0.426），PET主导性逐渐增强，且与独立成像流程的淀粉样SUVR（r=0.172，p=0.006）和FDG代谢生物标志物（r=-0.287，p=0.0005）验证一致。在1073名独立OASIS-3受试者中的外部复制确认了跨队列泛化性（H=166.99，p<0.0001，η²=0.156）。与SHAP的机制对比表明，基于消融的MCS捕捉了基于偏差的方法无法捕捉的临床有意义的模态依赖性。这些发现使MCNet成为个性化成像决策、临床试验分层及痴呆护理中可信AI的基础。
英文摘要
Multimodal neuroimaging combining structural MRI and positron emission tomography (PET) captures complementary structure-function relationships across the Alzheimer's disease (AD) continuum, yet existing artificial intelligence systems produce a single diagnostic label without quantifying which imaging modality drove that decision for a specific patient. We introduce the Modality Contribution Network (MCNet) and the Modality Contribution Score (MCS), the first per-patient attribution framework quantifying the shift in modality dominance from structural atrophy to amyloid and metabolic dysfunction across the cognitively normal to MCI to AD continuum. MCS is normalised to unity per subject via modality ablation (MCS_MRI_i + MCS_PET_i = 1.0 for every subject i), providing an interpretable, clinically actionable score that fluid biomarkers cannot supply. Applied to 327 ADNI-3 participants balanced across cognitively normal, mild cognitive impairment, and AD groups, MCNet achieved competitive three-class staging performance (AUC=0.881). The MCS revealed a statistically significant monotonic gradient (Kruskal-Wallis p<0.0001), with increasing PET dominance from cognitively normal (MCS_PET 0.412+/-0.229) through MCI (0.489+/-0.289) to AD (0.671+/-0.426), validated against amyloid SUVR (r=0.172, p=0.006) and FDG metabolic biomarkers (r=-0.287, p=0.0005) from separate imaging pipelines. External replication in 1,073 independent OASIS-3 subjects confirmed cross-cohort generalisability (H=166.99, p<0.0001, eta^2=0.156). A mechanistic comparison with SHAP demonstrated that ablation-based MCS captures clinically meaningful modality dependence that deviation-based methods cannot. These findings position MCNet as a foundation for personalised imaging decisions, clinical trial stratification, and trustworthy AI in dementia care.
Comments15 pages, 7 figures, Under review
