住区建成环境对老年人心理健康的影响及监测方法研究
王灵灵
摘 要
中国已成为全球老年人增速第一、老年人口最多的国家，人口老龄化将是今后一段时间的基本国情。老年人孤独抑郁、认知功能下降等心理问题凸显，提升其心理健康水平是实施积极应对人口老龄化国家战略的必然选择。住区建成环境是老年人的主要生活空间，越来越多研究证实住区建成环境对其心理健康有重要影响。然而，住区建成环境与老年人心理健康的复杂关系仍有待挖掘。因此，本文聚焦改善心理健康的住区建成环境设计和优化研究，开展如下工作:
1. 探究了多层次住区建成环境因素对老年人心理健康的影响机理。通过环境调研问卷，使用基于 Boruta 的住区建成环境因素选择方法识别出交通状况、绿地可及性、环境卫生状况、噪音、光线等 19 个对心理健康有重要影响的因素。进一步，提出了多尺度住区建成环境下健康行为-心理健康影响模型，研究了多层次社区层-住宅层环境因素、健康行为与心理健康间的关系，确定了社交活动、休闲锻炼是影响老年人心理健康，特别是认知和抑郁状况的重要中介变量。同时，针对室内噪音、光线关键影响因素，通过环境实验分析了物理环境因素耦合对个体主观和客观心理状态的影响，得到了面向心理健康的物理环境参数最佳阙值。基于以上研究结果，对养老基地建成环境实施改造，并提出了住区建成环境适应性优化策略。
2. 构建了面向心理健康异常评估的老年人居家行为监测量表。通过系统理论的推导，得到从居家行为活动序列和行为动作体征两个层次评估老年人心理健康状态。在居家行为活动序列方面，结合行为观察法和实际调查法，提出了面向心理健康的居家行为活动监测量表。从行为动作体征出发，基于全国老年人体检和健康调查数据，构建行为体征-心理健康关系模型，分析了行为动作体征与认知功能和抑郁状况的量化关系，识别出表征老年人心理健康风险的关键行为动作及其参数阅值。最终形成了老年人心理健康评估的居家行为监测量表，为老年人心理健康状态的早期识别提供了新的理论支撑。
3. 研发了基于老年人异常行为的心理健康状态监测技术，提供了一个居家环境下老年人健康状态监测的智能解决方案。通过构建健康监测技术接受度模型，获取了影响老年人健康监测技术使用的关键因素，进而提出居家环境下心理健康状态监测技术框架。具体地，使用多传感器获取老年人行为数据，构建Attention-CNN-LSTM模型和人体三维骨架信息的动作分类方法，以识别老年人动作异常。进一步基于人体长期时空行为数据提取个体关键行为模式，使用隐马尔可夫模型检测老年人的行为规律异常。动作异常和行为规律异常的识别结果能够支持评估老年人的心理健康状态。将上述技术集成到一个老年人健康状态预警系统中，测试表明该系统能通过识别老年人的异常行为有效实现对其异常心理健康状态的预警。该系统也被布设到全国多个养老基地用于监测老年人的健康状态。
本文的研究成果为适老性健康建成环境的发展提供了新的理论和方法支撑，对指导健康老龄化和健康城市建设具有重要意义。
关键词：住区建成环境；老年人心理健康；影响机制；监测技术；异常行为识别
Abstract
China has become the country with the highest growth rate of elderly people and the largest elderly population in the world, Population aging will be a fundamental national situation for a period of time. The mental health problems of loneliness, depression, and cognitive decline among the elderly have been increasingly prominent, and improving the mental health of the elderly is an inevitable choice for implementing the national strategy of actively responding to population aging. The residential built environment is the main living space for the elderly, and an increasing number of studies have confirmed its important role in the mental health of the elderly. However, the complex relationship between the residential built environment and the mental health of the elderly still requires further exploration. Therefore, this thesis focuses on the design and optimization of residential built-up environments for improving mental health The main contributions of this thesis are as follows:
l. This study explores the impact mechanism of multi-level residential built environment on the mental health of elderly people. Through an environmental survey questionnaire, a factor selection method for residential built environment based on Boruta is used to identify 19 factors that have a significant impact on human mental health. including traffic conditions, accessibility to green spaces, environmental hygiene, noise, and light. Furthermore, a relational model of health behavior and mental health in a multi-scale residential built environment is proposed. This model explores the interactive relationship between residential built environmental factors, health behavior, and mental health, and ultimately determines that social activities and leisure exercise are important mediating variables for the influence of built environmental factors on the cognitive state and depression status ofelderly people. In addition, for the key factors of indoor noise and light based on physical environment experiments, the coupling influence of sound and light environmental factors on human subjective and objective psychological states is clarified and the optimal threshold of physical environment parameters for residents' mental health has been determined. Based on the above research results, adaptive renovations of the built environment are then implemented in two retirement bases, and optimization strategies for residential built environment are derived
2. A home-based behavior monitoring scale for elderly people is constructed for the assessment of abnormal mental health status. Through the derivation of system theory, this study proposes to evaluate the mental health of the elderly from two behavioral levels: home behavior activity sequence and behavioral action signs. In terms of home behavior activity sequence, combined with the behavior observation method and the actual survey method,a home behavior activity monitoring scale for mental health is established. For behavioral action signs, based on the national physical examination and health survey data of the elderly. a behavioral action signs-mental health relationship model is constructed to analyze the quantitative relationship between behavioral action signs and cognitive scores and depression risk. Key behavioral action and their parameter thresholds that characterize the mental health risk of the elderly are identified. Finally, a home-based behavior monitoring scale for assessing the mental health of the elderly is formed to support early identification of mental health risks for the elderly in a home-based environment.
3. A monitoring technology for mental health status based on abnormal behavior is developed, which provides an intelligent solution for monitoring the health status of the elderly in a home environment. By investigating the acceptance of health monitoring technology, key factors affecting the use of health monitoring technology by elderly people are identified, thereby promoting the proposal of a technical framework for monitoring mental health status in home-based environments. Specifically, multiple sensors are used to collect behavioral data of the elderly, and an Attention-CNN-LSTM model and a action classification method using three-dimensional skeletal information are constructed to detect abnormal actions of the elderly. Furthermore, key behavioral patterns are extracted from long-term spatiotemporal behavioral data, and a hidden Markov model is utilized to generate long-term behavior sequences for detecting abnormal behavior patterns. The identification of abnormal actions and behavior patterns supports the detection of abnormal mental health status among the elderly. The above technologies are integrated into a set of elderly health status warning systems and testing shows that the system can effectively alerts abnormal mental health status among the elderly by identifying their abnormal behavior The system has also been deployed in multiple elderly care bases across the country to monitor the health status of the elderly.
The findings of this thesis provide new theoretical and methodological support for the development of age-friendly and health-promoting built environments, which are of great significance for guiding healthy aging and the construction of healthy cities.
Key words: Residential built environment; Mental helath of older adults; Impact mechanism; Monitoring technology, Abnormal behavior identification
国家数字建造技术
创新中心
