Lee Liz Yuanxi, Vaghari Delshad, Burkhart Michael C, Tino Peter, Montagnese Marcella, Li Zhuoyu, Zühlsdorff Katharina, Giorgio Joseph, Williams Guy, Chong Eddie, Chen Christopher, Underwood Benjamin R, Rittman Timothy, Kourtzi Zoe
Department of Psychology, University of Cambridge, Cambridge, CB2 3EB, United Kingdom.
School of Computer Science, University of Birmingham, Birmingham, B15 2TT, United Kingdom.
EClinicalMedicine. 2024 Jul 12;74:102725. doi: 10.1016/j.eclinm.2024.102725. eCollection 2024 Aug.
Predicting dementia early has major implications for clinical management and patient outcomes. Yet, we still lack sensitive tools for stratifying patients early, resulting in patients being undiagnosed or wrongly diagnosed. Despite rapid expansion in machine learning models for dementia prediction, limited model interpretability and generalizability impede translation to the clinic.
We build a robust and interpretable predictive prognostic model (PPM) and validate its clinical utility using real-world, routinely-collected, non-invasive, and low-cost (cognitive tests, structural MRI) patient data. To enhance scalability and generalizability to the clinic, we: 1) train the PPM with clinically-relevant predictors (cognitive tests, grey matter atrophy) that are common across research and clinical cohorts, 2) test PPM predictions with independent multicenter real-world data from memory clinics across countries (UK, Singapore).
PPM robustly predicts (accuracy: 81.66%, AUC: 0.84, sensitivity: 82.38%, specificity: 80.94%) whether patients at early disease stages (MCI) will remain stable or progress to Alzheimer's Disease (AD). PPM generalizes from research to real-world patient data across memory clinics and its predictions are validated against longitudinal clinical outcomes. PPM allows us to derive an individualized AI-guided multimodal marker (i.e. predictive prognostic index) that predicts progression to AD more precisely than standard clinical markers (grey matter atrophy, cognitive scores; PPM-derived marker: hazard ratio = 3.42, p = 0.01) or clinical diagnosis (PPM-derived marker: hazard ratio = 2.84, p < 0.01), reducing misdiagnosis.
Our results provide evidence for a robust and explainable clinical AI-guided marker for early dementia prediction that is validated against longitudinal, multicenter patient data across countries, and has strong potential for adoption in clinical practice.
Wellcome Trust, Royal Society, Alzheimer's Research UK, Alzheimer's Drug Discovery Foundation Diagnostics Accelerator, Alan Turing Institute.
早期预测痴呆症对临床管理和患者预后具有重大意义。然而，我们仍然缺乏能够早期对患者进行分层的敏感工具，这导致患者未被诊断或被误诊。尽管用于痴呆症预测的机器学习模型迅速扩展，但有限的模型可解释性和泛化性阻碍了其向临床的转化。
我们构建了一个强大且可解释的预测性预后模型（PPM），并使用真实世界、常规收集、非侵入性且低成本（认知测试、结构磁共振成像）的患者数据来验证其临床效用。为了提高可扩展性和对临床的泛化性，我们：1）使用在研究和临床队列中常见的与临床相关的预测指标（认知测试、灰质萎缩）训练PPM，2）使用来自多个国家（英国、新加坡）记忆门诊的独立多中心真实世界数据测试PPM的预测。
PPM能够可靠地预测（准确率：81.66%，AUC：0.84，敏感性：82.38%，特异性：80.94%）早期疾病阶段（轻度认知障碍）的患者是否会保持稳定或进展为阿尔茨海默病（AD）。PPM能够从研究数据推广到多个记忆门诊的真实世界患者数据，并且其预测通过纵向临床结果得到了验证。PPM使我们能够得出一个个性化的人工智能引导的多模态标志物（即预测性预后指数），该标志物比标准临床标志物（灰质萎缩、认知分数；PPM衍生标志物：风险比 = 3.42，p = 0.01）或临床诊断（PPM衍生标志物：风险比 = 2.84，p < 0.01）更精确地预测向AD的进展，从而减少误诊。
我们的结果为一种强大且可解释的临床人工智能引导的早期痴呆症预测标志物提供了证据，该标志物通过多个国家的纵向多中心患者数据得到了验证，并且在临床实践中具有很强的应用潜力。
惠康信托基金会、英国皇家学会、英国阿尔茨海默病研究协会、阿尔茨海默病药物发现基金会诊断加速器、艾伦·图灵研究所。
Zotero 插件
