可穿戴和便携式设备衍生的数字生物标志物在帕金森病前驱期及高危人群中的应用：一项范围综述

文献信息

中文摘要

背景： 帕金森病前驱期（pPD）是一个诊断前阶段，在此期间细微且异质性的变化可能难以通过常规临床评估发现。可穿戴或便携式设备衍生的数字生物标志物为客观和纵向评估提供了机会。本范围综述旨在梳理关于正式定义的pPD个体或帕金森病（PD）风险增高人群中数字生物标志物的现有证据，重点关注测量方法、分析方法、证据类型和方法学空白。
方法： 对PubMed、Web of Science、Embase、IEEE Xplore、Scopus和ACM Digital Library进行了系统检索，检索时间从数据库建库至2026年8月31日。纳入研究为在正式定义的pPD个体或PD风险增高人群中探讨可穿戴或便携式设备衍生数字生物标志物的研究。对研究特征、数字生物标志物领域、测量和分析方法以及证据类型进行了描述性综合。
结果： 共纳入43项研究，主要涉及iRBD个体或与α-突触核蛋白病相关的其他高危状态、基于人群或更广泛的前驱期风险队列、遗传高危人群以及特发性嗅觉减退/嗅觉缺失个体。数字生物标志物涵盖运动、言语、睡眠/活动和多领域特征，其中运动相关测量，尤其是步态和移动能力，被研究得最为广泛。机器学习方法在多个领域得到应用，而真正的多领域研究仍然有限。大多数研究提供了组间差异或判别性能的横断面证据，而关于未来PD诊断或表型转化的前瞻性证据较少。在测量方案、分析方法和验证策略方面观察到相当大的异质性。
结论： 可穿戴和便携式设备衍生的数字生物标志物是用于表征和监测前驱期PD风险谱个体的有前景的候选工具。然而，支持表型转化临床预测的证据仍受方法学异质性和缺乏足够纵向及外部验证的限制。未来研究应优先考虑标准化评估、大型多中心纵向队列、独立外部验证以及与已建立的临床和生物学标志物的整合。

英文摘要

BACKGROUND: Prodromal Parkinson's disease (pPD) represents a pre-diagnostic stage during which subtle and heterogeneous changes may be difficult to detect through conventional clinical assessments. Digital biomarkers derived from wearable or portable devices offer opportunities for objective and longitudinal assessment. This scoping review aimed to map the current evidence on digital biomarkers in individuals with formally defined pPD or populations at increased risk of Parkinson's disease (PD), focusing on measurement approaches, analytical methods, types of evidence, and methodological gaps.
METHODS: A systematic search of PubMed, Web of Science, Embase, IEEE Xplore, Scopus, and ACM Digital Library was conducted from database inception to August 31, 2026. Studies investigating digital biomarkers derived from wearable or portable devices in individuals with formally defined pPD or populations at increased risk of PD were included. Study characteristics, digital biomarker domains, measurement and analytical approaches, and types of evidence were synthesized descriptively.
RESULTS: Forty-three studies were included, primarily involving individuals with iRBD or other high-risk states associated with α-synucleinopathies, population-based or broader prodromal-risk cohorts, genetic high-risk populations, and individuals with idiopathic hyposmia/anosmia. Digital biomarkers encompassed motor, speech, sleep/activity, and multidomain features, with motor-related measures, particularly gait and mobility, being the most extensively investigated. Machine-learning approaches were applied across several domains, whereas genuinely multidomain studies remained limited. Most studies provided cross-sectional evidence of group differences or discriminative performance, whereas prospective evidence regarding future PD diagnosis or phenoconversion was less extensive. Considerable heterogeneity was observed in measurement protocols, analytical approaches, and validation strategies.
CONCLUSION: Wearable- and portable-device-derived digital biomarkers represent promising candidate tools for characterizing and monitoring individuals across the prodromal PD risk spectrum. However, evidence supporting clinical prediction of phenoconversion remains limited by methodological heterogeneity and a lack of sufficient longitudinal and external validation. Future research should prioritize standardized assessment, large, multicenter longitudinal cohorts, independent external validation, and integration with established clinical and biological markers.