创伤性脑损伤患者ICU死亡与不良结局的预测:一项基于多中心大数据的机器学习研究

注册号:

Registration number:

ChiCTR2500115382 

最近更新日期:

Date of Last Refreshed on:

2025-12-25 10:40:39 

注册时间:

Date of Registration:

2025-12-25 00:00:00 

注册号状态:

预注册

Registration Status:

Prospective registration

注册题目:

创伤性脑损伤患者ICU死亡与不良结局的预测:一项基于多中心大数据的机器学习研究

Public title:

Predicting ICU Mortality and Adverse Outcomes in Traumatic Brain Injury Patients: A Machine Learning Study Based on Multicenter Big Data

注册题目简写:

English Acronym:

研究课题的正式科学名称:

创伤性脑损伤患者ICU死亡与不良结局的预测:一项基于多中心大数据的机器学习研究

Scientific title:

Predicting ICU Mortality and Adverse Outcomes in Traumatic Brain Injury Patients: A Machine Learning Study Based on Multicenter Big Data

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:

王盈 

研究负责人:

王伯栋 

Applicant:

Wang Ying  

Study leader:

Wang Bodong  

申请注册联系人电话:

Applicant telephone:

+86 152 6566 8819

研究负责人电话:

Study leader's
telephone:

+86 198 6181 1291

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:

cywy0624@163.com

研究负责人电子邮件:

Study leader's E-mail:

bdwang_neurosurg@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:

中国山东省济南市天桥区师范路25号

研究负责人通讯地址:

中国山东省济南市天桥区师范路25号

Applicant address:

25 Shifan road, Tianqiao district, Jinan, Shandong, China

Study leader's address:

25 Shifan road, Tianqiao district, Jinan, Shandong, China

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:

中国人民解放军联勤保障部队第九六〇医院

Applicant's institution:

The 960th Hospital of the PLA Joint Logistics Support Force

研究负责人所在单位:

中国人民解放军联勤保障部队第九六〇医院

Affiliation of the Leader:

The 960th Hospital of the PLA Joint Logistics Support Force

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:

(2025)科研伦理审第(215)号

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:

解放军第九六〇医院科研伦理委员会

Name of the ethic committee:

The 960th Hospital of the PLA Joint Logistics Support Force Research Ethics Committee

伦理委员会批准日期:

Date of approved by ethic committee:

2025-10-28 00:00:00

伦理委员会联系人:

孙志东

Contact Name of the ethic committee:

Sun Zhidong

伦理委员会联系地址:

中国山东省济南市天桥区师范路25号

Contact Address of the ethic committee:

25 Shifan road, Tianqiao district, Jinan, Shandong China

伦理委员会联系人电话:

Contact phone of the ethic committee:

+86 531 5166 6145

伦理委员会联系人邮箱:

Contact email of the ethic committee:

研究实施负责(组长)单位:

中国人民解放军联勤保障部队第九六〇医院

Primary sponsor:

The 960th Hospital of the PLA Joint Logistics Support Force

研究实施负责(组长)单位地址:

中国山东省济南市天桥区师范路25号

Primary sponsor's address:

25 Shifan road, Tianqiao district, Jinan, Shandong, China

试验主办单位(项目批准或申办者):

Secondary sponsor:

国家:

中国

省(直辖市):

山东

市(区县):

济南

Country:

China

Province:

Shandong

City:

Jinan

单位(医院):

中国人民解放军联勤保障部队第九六〇医院

具体地址:

山东省济南市天桥区师范路25号

Institution
hospital:

The 960th Hospital of the PLA Joint Logistics Support Force

Address:

25 Shifan road, Tianqiao district, Jinan, Shandong, China

经费或物资来源:

山东省卫生健康科创团队建设项目

Source(s) of funding:

Shandong Provincial Health Science and Technology Innovation Team Development Project

研究疾病:

创伤性脑损伤  

Target disease:

Traumatic brain injury

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:

队列研究 

Study design:

Cohort study 

研究目的:

本研究的总体目的是充分利用多中心大数据资源,开发并验证一套能够综合预测TBI患者ICU 30天死亡率及VAP、AKI、DVT、GIB四种关键并发症的机器学习模型。该研究致力于通过整合国际公开数据库与国内真实临床数据,确保模型兼具高预测精度、良好的泛化能力以及临床可解释性。具体目标包括系统性地完成多源数据的整合与预处理以构建高质量数据集;应用多种先进算法分别构建五项结局的预测模型并进行严格的内部验证与优化;在独立的国内多中心数据上对优选模型进行外部验证以评估其稳健性;并利用可解释性人工智能技术揭示模型的决策依据,识别关键风险因素。最终,本研究旨在将理论模型转化为潜在的临床决策支持工具,为早期识别高危患者、实施个体化干预、合理分配医疗资源提供科学依据,从而有效改善患者预后,减轻社会医疗负担。  

Objectives of Study:

The overarching objective of this study is to leverage multicenter big data resources to develop and validate a set of machine learning models capable of comprehensively predicting 30-day ICU mortality and four critical complications—VAP, AKI, DVT, and GIB—in patients with traumatic brain injury (TBI). By integrating international public databases with real-world domestic clinical data, this research aims to ensure that the models exhibit high predictive accuracy, robust generalization capability, and clinical interpretability. Specific objectives include systematically integrating and preprocessing multi-source data to construct a high-quality dataset; applying multiple advanced algorithms to develop predictive models for the five outcomes and conducting rigorous internal validation and optimization; performing external validation of the selected optimal models using an independent domestic multicenter dataset to assess their robustness; and utilizing explainable artificial intelligence techniques to untangle the decision-making rationale of the models and identify key risk factors. Ultimately, this study seeks to translate theoretical models into potential clinical decision support tools, providing a scientific basis for the early identification of high-risk patients, the implementation of individualized interventions, and the rational allocation of medical resources, thereby effectively improving patient outcomes and alleviating the societal healthcare burden.

药物成份或治疗方案详述:

无 

Description for medicine or protocol of treatment in detail:

None 

纳入标准:

Inclusion criteria

排除标准:

1.非首次ICU入院或数据缺失严重者 2.合并严重全身创伤或多发系统疾病干扰预后判断 3.妊娠妇女

Exclusion criteria:

1.Non-first ICU admission or severe data deficiency 2.Coexistence of severe systemic trauma or multiple system diseases that interfere with prognosis assessment 3.Pregnancy

研究实施时间:

Study execute time:

From 2026-02-15 00:00:00 To 2026-12-31 00:00:00  

征募观察对象时间:

Recruiting time:

From 2026-02-15 00:00:00 To 2026-12-31 00:00:00

干预措施:

Interventions:

组别:

观察组

样本量:

2000

Group:

Observational 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):

山东 

市(区县):

 

Country:

China

Province:

Shandong

City:

单位(医院):

中国人民解放军联勤保障部队第九六〇医院  

单位级别:

三甲 

Institution
hospital:

The 960th Hospital of the PLA Joint Logistics Support Force

Level of the institution:

Tertiary A

国家:

中国

省(直辖市):

陕西 

市(区县):

 

Country:

China

Province:

Shanxi

City:

单位(医院):

空军军医大学唐都医院 

单位级别:

三甲 

Institution
hospital:

Air Force Medical University Tangdu Hospital

Level of the institution:

Tertiary A

测量指标:

Outcomes:

指标中文名:

30天死亡

指标类型:

主要指标

Outcome:

30-day mortality

Type:

Primary indicator

测量时间点:

ICU入院后第30天

测量方法:

通过电子病历系统记录的院内死亡信息,结合出院随访记录或死亡登记系统确认30天生存状态。

Measure time point of outcome:

30 days after ICU admission

Measure method:

All-cause mortality was determined from electronic medical records, supplemented by discharge follow-up records or death registry data.

指标中文名:

急性肾损伤

指标类型:

次要指标

Outcome:

Acute kidney injury

Type:

Secondary indicator

测量时间点:

ICU入院后0–7天及住院期间

测量方法:

根据KDIGO标准,结合血清肌酐变化、尿量记录及病历诊断信息判定。

Measure time point of outcome:

Within 0–7 days after ICU admission and throughout ICU stay

Measure method:

AKI was defined according to KDIGO criteria using serum creatinine changes, urine output records, and documented clinical diagnoses.

指标中文名:

消化道出血

指标类型:

次要指标

Outcome:

Gastrointestinal bleeding

Type:

Secondary indicator

测量时间点:

ICU住院期间

测量方法:

通过病历中明确记录的消化道出血事件,结合内镜检查、血红蛋白下降及止血或输血治疗记录判定。

Measure time point of outcome:

During ICU stay

Measure method:

GIB was identified based on documented gastrointestinal bleeding events, endoscopic findings, hemoglobin decrease, and hemostatic or transfusion records.

指标中文名:

呼吸相关性肺炎

指标类型:

次要指标

Outcome:

Pneumonia

Type:

Secondary indicator

测量时间点:

ICU住院期间,机械通气>=48小时后

测量方法:

依据临床诊断记录、ICD编码,结合影像学检查及感染相关实验室指标综合判定。

Measure time point of outcome:

During ICU stay ,>=48 hours after initiation of mechanical ventilation

Measure method:

VAP was identified based on clinical diagnosis records, ICD codes, imaging findings, and infection-related laboratory indicators.

指标中文名:

静脉血栓栓塞

指标类型:

次要指标

Outcome:

Venous thromboembolism

Type:

Secondary indicator

测量时间点:

ICU住院期间

测量方法:

依据彩色多普勒超声检查结果、临床诊断记录及ICD编码确认。

Measure time point of outcome:

During ICU stay

Measure method:

DVT was confirmed by Doppler ultrasound findings, clinical diagnosis records, and ICD codes.

采集人体标本:

Collecting sample(s)
from participants:

标本中文名:

血液

组织:

Sample Name:

Blood

Tissue:

人体标本去向

使用后销毁  

说明

Fate of sample:

Destruction after use  

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 18 years
最大 Max age 100 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):

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

None

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:

None

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

是Yes

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):

数据在文章发表后可以共享,去标识化后的数据可以在合理请求下,从通讯作者处获取。

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

The original raw data will be made publicly available after publica. However, de-identified data can be obtained from the corresponding author upon reasonable request

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:

电子采集和管理系统

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

Electronic Data Capture, EDC

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:

 2025-12-25 10:40:35