机器学习构建社区老年人跌倒风险预测模型

注册号:

Registration number:

ChiCTR2500104155 

最近更新日期:

Date of Last Refreshed on:

2025-06-12 08:28:28 

注册时间:

Date of Registration:

2025-06-12 00:00:00 

注册号状态:

预注册

Registration Status:

Prospective registration

注册题目:

机器学习构建社区老年人跌倒风险预测模型

Public title:

Machine Learning to Build a Fall Risk Prediction Model for Community-Dwelling Older Adults

注册题目简写:

English Acronym:

研究课题的正式科学名称:

机器学习构建社区老年人跌倒风险预测模型

Scientific title:

Development and Validation of a Machine Learning–Based Fall Risk Prediction Model for Community-Dwelling Older Adults

研究课题代号(代码):

Study subject ID:

在二级注册机构或其它机构的注册号:

The registration number of the Partner Registry or other register:

申请注册联系人:

胡海洲 

研究负责人:

胡海洲 

Applicant:

Haizhou Hu 

Study leader:

Haizhou Hu 

申请注册联系人电话:

Applicant telephone:

+86 188 1171 3698

研究负责人电话:

Study leader's
telephone:

+86 188 1171 3698

申请注册联系人传真 :

Applicant Fax:

研究负责人传真:

Study leader's fax:

申请注册联系人电子邮件:

Applicant E-mail:

18811713698@163.com

研究负责人电子邮件:

Study leader's E-mail:

18811713698@163.com

申请单位网址(自愿提供):

Applicant website(voluntary supply):

研究负责人网址(自愿提供):

Study leader's website(voluntary supply):

申请注册联系人通讯地址:

北京市海淀区信息路48号

研究负责人通讯地址:

北京市海淀区信息路48号

Applicant address:

No.48,Information Road, Haidian District, Beijing

Study leader's address:

No.48,Information Road, Haidian District, Beijing

申请注册联系人邮政编码:

Applicant postcode:

研究负责人邮政编码:

Study leader's postcode:

申请人所在单位:

北京体育大学

Applicant's institution:

Beijing Sport University

研究负责人所在单位:

北京体育大学

Affiliation of the Leader:

Beijing Sport University

是否获伦理委员会批准:

Approved by ethic committee:

Yes

伦理委员会批件文号:

Approved No. of ethic committee:

2025254H

伦理委员会批件附件:

Approved file of Ethical Committee:

查看附件View

批准本研究的伦理委员会名称:

北京体育大学运动科学实验伦理委员会

Name of the ethic committee:

Beijing Sport University Ethics Committee for Sports Science Experiments

伦理委员会批准日期:

Date of approved by ethic committee:

2025-06-11 00:00:00

伦理委员会联系人:

梅涛

Contact Name of the ethic committee:

Tao Mei

伦理委员会联系地址:

北京市海淀区信息路48号

Contact Address of the ethic committee:

No.48,Information Road, Haidian District, Beijing

伦理委员会联系人电话:

Contact phone of the ethic committee:

+86 62989306

伦理委员会联系人邮箱:

Contact email of the ethic committee:

18811713698@163.com

研究实施负责(组长)单位:

北京体育大学

Primary sponsor:

Beijing Sport University

研究实施负责(组长)单位地址:

北京市海淀区信息路48号

Primary sponsor's address:

No.48,Information Road, Haidian District, Beijing

试验主办单位(项目批准或申办者):

Secondary sponsor:

国家:

中国

省(直辖市):

北京

市(区县):

海淀区

Country:

China

Province:

Beijing

City:

Haidian District

单位(医院):

北京体育大学

具体地址:

北京市海淀区信息路48号

Institution
hospital:

Beijing Sport University

Address:

No.48,Information Road,Haidian District,Beijing

经费或物资来源:

自筹

Source(s) of funding:

Self-raised funds

研究疾病:

老年人跌倒  

Target disease:

Elderly falls

研究疾病代码:

Target disease code:

研究类型:

观察性研究

Study type:

Observational study

研究所处阶段:

其它 

Study phase:

N/A

研究设计:

横断面 

Study design:

Cross-sectional 

研究目的:

对老年人跌倒风险进行准确预测,及时制订针对性的防跌倒干预措施,是减少老年人跌倒发生率的有效策略。然而,目前尚缺乏社区老年人基于双任务下步态姿势控制生物力学特征指标、跌倒风险相关测评指标和认知功能相关指标预测跌倒风险的研究。此外,使用机器学习对老年人跌倒风险的预测能力优于传统预测模型,但不同算法各有特点,比较不同算法建立的模型对于提高老年人跌倒风险预测的准确性具有重要意义。综上所述,本研究目的是分析基于复杂双任务下社区跌倒老年人步态姿势特征,确立双任务下能够区分社区老年人跌倒的步态姿势控制生物力学参数、跌倒风险相关测评指标和认知功能相关指标,使用神经网络、极端梯度提升、随机森林、朴素贝叶斯、逻辑回归、支持向量机6种机器学习算法构建社区老年人跌倒风险预测模型,对比找到最佳预测老年人跌倒的机器学习预测模型,采用SHAP方法增强机器学习模型的可解释性。本研究旨在帮助医护人员及早识别跌倒个体并及时采取针对性干预措施,降低社区老年人跌倒风险。翻译成专业的英文,符合英文期刊的标准  

Objectives of Study:

Accurately predicting the risk of falls in the elderly and promptly developing targeted fall prevention interventions are effective strategies to reduce the incidence of falls among older adults. However, there is currently a lack of research on predicting fall risk in community-dwelling elderly individuals based on biomechanical characteristics of gait posture control under dual-task conditions, fall risk-related assessment indicators, and cognitive function-related indicators. Additionally, machine learning demonstrates superior predictive capabilities for fall risk in the elderly compared to traditional predictive models. However, different algorithms have unique characteristics, and comparing models built with various algorithms is of significant importance for improving the accuracy of fall risk prediction in older adults. In summary, the objective of this study is to analyze the gait posture characteristics of community-dwelling elderly individuals who have experienced falls under complex dual-task conditions, identify biomechanical parameters of gait posture control, fall risk-related assessment indicators, and cognitive function-related indicators that can differentiate falls in community-dwelling elderly individuals under dual-task conditions. Six machine learning algorithms—neural networks, extreme gradient boosting, random forests, naive Bayes, logistic regression, and support vector machines—will be used to construct fall risk prediction models for community-dwelling elderly individuals. The best machine learning prediction model for fall risk in the elderly will be identified through comparison, and the SHAP method will be employed to enhance the interpretability of the machine learning models. This study aims to assist healthcare professionals in early identification of individuals at risk of falls and the timely implementation of targeted interventions to reduce fall risk among community-dwelling elderly individuals.

药物成份或治疗方案详述:

 

Description for medicine or protocol of treatment in detail:

 

纳入标准:

Inclusion criteria

排除标准:

排除标准:(1)有严重的心血管疾病、呼吸系统疾病、神经退行性疾病、视觉障碍或其他严重疾病,可能会影响平衡或移动能力或受试者安全;(2)诊断为中度或重度认知障碍,无法理解或执行测试任务;(3)过去六个月内经历过重大手术或严重外伤,可能影响平衡和活动能力;(4)正在服用影响平衡或认知功能的药物,如某些类型的精神药物或镇静剂;(5)一年内发生过因车祸、外部暴力、急性疾病(如中风或心脏病发作)等非源性原因导致的跌倒;(6)当前没有参与任何药理学研究;(7)声明不愿参加研究

Exclusion criteria:

Exclusion Criteria: (1) Having severe cardiovascular disease, respiratory disease, neurodegenerative disease, visual impairment, or other serious conditions that may affect balance or mobility or pose a risk to the subject's safety; (2) Being diagnosed with moderate or severe cognitive impairment, unable to understand or perform test tasks; (3) Having undergone major surgery or severe trauma within the past six months, which may affect balance and mobility; (4) Currently taking medications that affect balance or cognitive function, such as certain types of psychotropic drugs or sedatives; (5) Having experienced a fall due to non-iatrogenic causes (e.g., car accidents, external violence, acute illnesses such as stroke or heart attack) within the past year; (6) Currently not participating in any pharmacological research; (7) Declaring unwillingness to participate in the study

研究实施时间:

Study execute time:

From 2025-06-15 00:00:00 To 2025-12-31 00:00:00  

征募观察对象时间:

Recruiting time:

From 2025-06-15 00:00:00 To 2025-07-10 00:00:00

干预措施:

Interventions:

组别:

跌倒组

样本量:

60

Group:

Faller group

Sample size:

干预措施:

干预措施代码:

Intervention:

None

Intervention code:

组别:

未跌倒组

样本量:

80

Group:

Non-faller group

Sample size:

干预措施:

干预措施代码:

Intervention:

None

Intervention code:

研究实施地点:

Countries of recruitment and research settings:

国家:

中国

省(直辖市):

北京 

市(区县):

海淀区 

Country:

China

Province:

Beijing

City:

Haidian District

单位(医院):

北京体育大学 

单位级别:

大学 

Institution
hospital:

Beijing Sport University

Level of the institution:

University

测量指标:

Outcomes:

指标中文名:

步态变异性:步长,步宽,步速,步频,步态周期,支撑期百分比,摆动期百分比,双支撑期百分比

指标类型:

主要指标

Outcome:

Gait variability parameters: Step length Step width Gait speed Cadence (steps per minute) Gait cycle duration Percentage of stance phase Percentage of swing phase Percentage of double-support phase

Type:

Primary indicator

测量时间点:

测量方法:

Measure time point of outcome:

Measure method:

指标中文名:

步态不对称指数

指标类型:

次要指标

Outcome:

Gait Asymmetry Index

Type:

Secondary indicator

测量时间点:

测量方法:

Measure time point of outcome:

Measure method:

采集人体标本:

Collecting sample(s)
from participants:

标本中文名:

组织:

Sample Name:

None

Tissue:

人体标本去向

其它  

说明

Fate of sample:

0thers  

Note:

征募研究对象情况:

Recruiting status:

尚未开始

Not yet recruiting

年龄范围:

Participant age:

最小 Min age 65 years
最大 Max age years

性别:

男女均可

Gender:

Both

随机方法(请说明由何人用什么方法产生随机序列):

随机分配序列由一名独立统计学家使用SAS v9.4软件中的PROC PLAN过程生成,采用区组随机化方案(区组大小=4)。生成的序列置于按顺序编号、不透明且密封的信封中,由一名研究助理负责准备。每位受试者完成基线评估后,由研究协调员按编号打开相应信封,进行分组分配。

Randomization Procedure (please state who generates the random number sequence and by what method):

The random allocation sequence was generated by an independent statistician using the PROC PLAN procedure in SAS v9.4, employing a block‐randomization scheme (block size = 4). The sequence was placed into sequentially numbered, opaque, sealed envelopes prepared by a research assistant. After each participant completed baseline assessments, the study coordinator opened the next envelope in order to assign the participant to their group.

是否公开试验完成后的统计结果:

Calculated Results after the Study Completed public access:

公开/Public

盲法:

Blinding:

试验完成后的统计结果(上传文件):

Calculated Results after
the Study Completed(upload file):

是否共享原始数据:

IPD sharing

否No

共享原始数据的方式(说明:请填入公开原始数据日期和方式,如采用网络平台,需填该网络平台名称和网址):

本研究原始数据也许会在2025年12月31日通过Open Science Framework (OSF)平台(https://osf.io/xxxxx)公开共享,届时研究者可免费下载和使用,是否公开将和团队商定后再确定。

The way of sharing IPD”(include metadata and protocol, If use web-based public database, please provide the url):

The raw data of this study may be made publicly available via the Open Science Framework (OSF) platform (https://osf.io/xxxxx) on December 31, 2025, at which point researchers will be able to download and use it free of charge. The final decision on whether to share will be made in consultation with the project team.

数据采集和管理(说明:数据采集和管理由两部分组成,一为病例记录表(Case Record Form, CRF),二为电子采集和管理系统(Electronic Data Capture, EDC),如ResMan即为一种基于互联网的EDC:

病例记录表(Case Record Form, CRF) 本研究制定专用纸质CRF,用于记录各项研究内容,包括受试者人口学信息、基线评估结果、步态生物力学参数、双任务测试数据、不良事件等。 由经过培训的研究人员按照标准操作流程(SOP)填写,确保内容完整、准确。 电子数据采集与管理系统(Electronic Data Capture, EDC) 所有纸质CRF数据将录入ResMan互联网EDC系统。 ResMan系统具备: 自动校验功能:包括范围校验、一致性校验和必填字段提醒; 电子查询管理:实时生成、跟踪和关闭查询; 审计追踪:记录数据录入、修改及用户操作日志; 权限分级:根据角色分配阅读和编辑权限,保障数据安全。 系统每日自动备份数据,并存储于安全服务器,符合GCP和数据管理规范。 此“纸质CRF → ResMan EDC”双层管理模式,可有效提升数据完整性、准确性及可追溯性,为研究质量提供可靠保障。

Data collection and Management (A standard data collection and management system include a CRF and an electronic data capture:

Case Record Form (CRF) A dedicated paper CRF has been developed to capture all study-specific information, including participant demographics, baseline assessment results, gait biomechanical parameters, dual-task performance data, and adverse events. Trained study personnel complete the CRFs in accordance with Standard Operating Procedures (SOPs) to ensure completeness and accuracy. Electronic Data Capture and Management System (EDC) All data from the paper CRFs are entered into the ResMan web-based EDC system. ResMan provides: Automated validation checks (range checks, consistency checks, and required-field prompts) Electronic query management, enabling real-time generation, tracking, and resolution of data queries Audit trails, logging all data entries, modifications, and user activities Role-based access control, ensuring that only authorized users can view or edit data The system performs daily automatic backups to secure servers, in full compliance with GCP and data management guidelines. This two-tiered "Paper CRF → ResMan EDC" approach enhances data integrity, accuracy, and traceability, thereby providing a robust guarantee of study quality.

数据与安全监察委员会:

Data and Safety Monitoring Committee:

暂未确定/Not yet

注册人:

Name of Registration:

 2025-06-12 08:28:23