DOI: 10.66106/lyqaav.20250101 ISSN: 3105-6873

可穿戴设备与AI算法在老年衰弱综合征动态监测中的应用（Application of wearable devices and AI algorithm in dynamic monitoring of senile asthenia syndrome）

张明昭 Mingzhao Zhang
Show PDF Cite
Abstract：The degree of population aging continues to deepen, and geriatric asthenia syndrome has become a core issue in the fi eld of geriatrics. Weakness is not a single organ damage, but a dynamic change process caused by the decline of multi system functional reserve. Traditional evaluation methods rely on regular screening, which can not accurately capture this fl uctuation. The emergence of wearable devices and artifi cial intelligence algorithms has opened up a new direction for the continuous monitoring of weakness. Related studies have been able to preliminarily collect and analyze signals such as gait, heart rate, activity, etc., but there are still some major shortcomings: most of the existing monitoring are only for a single dimension or short-term state, and the overall presentation of the weak dynamic trajectory is not achieved; The algorithm model is mostly based on cross-sectional data, which is diffi cult to adapt to the nonlinear changes of individuals over time; The data collected by the equipment and the clinical evaluation standards are not well connected, which reduces the practical value of the monitoring results. These problems arise because the occurrence and development of weakness are related to multiple factors such as physiology, psychology and society. The current technical means have not yet integrated these dimensions, coupled with the lack of a unifi ed data specifi cation and dynamic modeling framework, so that most wearable devices can only display data and cannot be transformed into clinical decision support. In this paper, we propose a multi parameter dynamic perception framework for elderly weakness monitoring. Wearable devices are used to continuously collect multi-source information such as physiological rhythm, daily activities and behavior characteristics. The weak state and its migration trend are identifi ed by interpretable time series algorithm, and the individualized risk stratifi cation and early warning path are built combined with the comprehensive evaluation index of the elderly. The framework changes the assessment mode from static screening to dynamic tracking, and integrates the cognition of geriatrics on the heterogeneity of functional decline into the technical design.

More from our Archive