ChiCTR2500115619 版本V1.0 版本创建时间2025/12/29 15:41:49 中国临床试验注册中心

审核状态:

Project audit state:

通过审核

Successful

注册号:

Registration number:

ChiCTR2500115619 

最近更新日期:

Date of Last Refreshed on:

2025-12-29 15:41:21 

注册时间:

Date of Registration:

2025-12-29 00:00:00 

注册号状态:

预注册

Registration Status:

Prospective registration

注册题目:

赋能罕见病诊断:一种基于合成数据驱动推理性大语言模型,对应长尾医疗挑战

Public title:

Empowering Rare Disease Diagnosis: A Synthetic Data-Driven Reasoning LLM for Long-Tail Medical Challenge

注册题目简写:

大语言模型辅助罕见病诊断

English Acronym:

Large language model facilitated rare disease diagnosis

研究课题的正式科学名称:

赋能罕见病诊断:一种基于合成数据驱动推理性大语言模型,对应长尾医疗挑战

Scientific title:

Empowering Rare Disease Diagnosis: A Synthetic Data-Driven Reasoning LLM for Long-Tail Medical Challenge

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:

陈海潮 

研究负责人:

Wong Tien Yin 

Applicant:

Haichao Chen 

Study leader:

Wong Tien Yin 

申请注册联系人电话:

Applicant telephone:

+86 181 1630 8936

研究负责人电话:

Study leader's
telephone:

+86 151 0100 3732

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:

chc1998.12138@gmail.com

研究负责人电子邮件:

Study leader's E-mail:

wongtienyin@tsinghua.edu.cn

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:

北京东城区赵堂子胡同5号

研究负责人通讯地址:

中国北京市海淀区双清路30号

Applicant address:

No. 5 Zhaotangzi Hutong, Dongcheng District, Beijing

Study leader's address:

No. 30 Shuangqing Road, Haidian District, Beijing, 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:

清华大学,协和医院

Applicant's institution:

Tsinghua Universiy, Peking Union Medical Colldge Hospital

研究负责人所在单位:

清华大学

Affiliation of the Leader:

Tsinghua Universiy

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:

THU-01-2025-1070

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:

清华大学科技伦理委员会

Name of the ethic committee:

Tsinghua University Science and Technology Ethics Committee (Medicine)

伦理委员会批准日期:

Date of approved by ethic committee:

2025-11-18 00:00:00

伦理委员会联系人:

Yandong Zhu

Contact Name of the ethic committee:

Yandong Zhu

伦理委员会联系地址:

中国北京市海淀区双清路30号

Contact Address of the ethic committee:

No. 30 Shuangqing Road, Haidian District, Beijing, China

伦理委员会联系人电话:

Contact phone of the ethic committee:

+86 10 6279 6175

伦理委员会联系人邮箱:

Contact email of the ethic committee:

yandong2012@mail.tsinghua.edu.cn

研究实施负责(组长)单位:

清华大学

Primary sponsor:

Tsinghua University

研究实施负责(组长)单位地址:

中国北京市海淀区双清路30号

Primary sponsor's address:

No. 30 Shuangqing Road, Haidian District, Beijing, 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):

北京

市(区县):

Country:

China

Province:

Beijing

City:

单位(医院):

清华大学

具体地址:

中国北京市海淀区双清路30号

Institution
hospital:

Tsinghua University

Address:

No. 30 Shuangqing Road, Haidian District, Beijing, China

经费或物资来源:

清华大学

Source(s) of funding:

Tsinghua University

研究疾病:

无  

Target disease:

None

研究疾病代码:

Target disease code:

研究类型:

干预性研究

Study type:

Interventional study

研究所处阶段:

其它 

Study phase:

N/A

研究设计:

随机平行对照 

Study design:

Parallel 

研究目的:

罕见病诊断是一个关键的公共卫生挑战。全球范围内,罕见病诊断延迟平均长达 4–5 年,这给患者及其家庭带来了沉重负担。造成这一问题的重要原因之一,是临床医生对数以千计的罕见病认知与熟悉程度有限。传统方法(例如基于表型在互联网上检索)虽可在一定程度上补充临床知识,但支持作用仍较为有限。我们团队近期开发了专门的 RaDaR 模型,在罕见病诊断基准测试以及回顾性真实世界队列中均显示出显著提升的优异表现。  

Objectives of Study:

Rare disease diagnosis represents a critical public health challenge. Diagnostic delay—averaging 4–5 years globally—places a substantial burden on patients and their families. A major contributing factor is the limited awareness and familiarity among physicians regarding the thousands of rare diseases. Traditional approaches, such as search on internet by phenotype, can supplement clinical knowledge but provide only limited support. Our group’s specialized RaDaR model has been developed recently, demonstrating improved remarkable performance in both rare disease diagnostic benchmark and retrospective real-world cohorts.

药物成份或治疗方案详述:

 

Description for medicine or protocol of treatment in detail:

 

纳入标准:

1. 通过医师执业资格考试的医生; 2. 内科、儿科和神经内科的医生; 3. 目前在门诊工作的医生; 4. 完成知情同意书的医生。

Inclusion criteria

1. physicians that pass the physician licensing exam; 2. physicians in in internal medicine, pediatrics, and neurology; 3. Physicians that are currently working in clinics 4. Physicians that finish the informed consent form.

排除标准:

1. 来自其他科室的医生; 2. 已退出临床工作的医生; 3. 不愿意完成知情同意书的医生。

Exclusion criteria:

1. Physicians that from other departments; 2. Physicians that has quitted from clinics work; 3. Physicians that are unwilling to finish the informed consent form.

研究实施时间:

Study execute time:

From 2026-01-01 00:00:00 To 2026-01-15 00:00:00  

征募观察对象时间:

Recruiting time:

From 2026-01-01 00:00:00 To 2026-01-05 00:00:00

干预措施:

Interventions:

组别:

干预组

样本量:

35

Group:

Interventional group

Sample size:

干预措施:

可使用大型语言模型(RaDaR)和网络搜索功能

干预措施代码:

Intervention:

With access to large language model and web search

Intervention code:

组别:

控制组

样本量:

35

Group:

Control group

Sample size:

干预措施:

仅使用网络搜索功能

干预措施代码:

Intervention:

With access to web search only

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):

 

Country:

China

Province:

Beijing

City:

单位(医院):

清华大学 

单位级别:

大学 

Institution
hospital:

Tsinghua University

Level of the institution:

University

测量指标:

Outcomes:

指标中文名:

鉴别诊断列表的准确性(即最终诊断列表是否包含已确诊的罕见病诊断)

指标类型:

主要指标

Outcome:

The accuracy of the differential diagnosis list (whether the final diagnosis list include the confirmed rare disease diagnosis)

Type:

Primary indicator

测量时间点:

测量方法:

专家评估

Measure time point of outcome:

Measure method:

Evaluation from human expert

指标中文名:

受试者对帮助性的主观评价

指标类型:

次要指标

Outcome:

The subjective evalation of the helpfulness

Type:

Secondary indicator

测量时间点:

测量方法:

自我报告

Measure time point of outcome:

Measure method:

self reported

指标中文名:

诊断时间

指标类型:

次要指标

Outcome:

Time for the diagnosis

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 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):

随机分组采用分层区组分配程序,区组大小为 2。一名独立研究人员将通过集中随机化系统协助完成分组分配。参与者将按科室(内科、儿科、神经内科)进行分层。分配顺序将按照参与者完成基线调查的先后顺序进行。为尽量降低招募人员的知晓程度,随机分配序列由 S.H. 保管,并对负责招募的人员进行盲法处理。

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

Randomization followed a stratified allocation procedure with block size 2. An independent researcher will help allocate the group using a centralized randomization system. Participants will be stratified by department (internal medicine, pediatrics, neurology). The allocation order will follow the order when they finish the baseline survey. To minimize the awareness of the recruiting, the allocation series was kept by S.H. and masked to the people recruiting.

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:

评估两组答案的人类专家对分组分配情况实行盲法(不知道参与者所属组别)

Blinding:

The humen experts who evluated the answers of two groups were bilindedd of the group assignment.

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):

个体参与者数据(IPD)将不予共享,但统计结果将以发表、报告或展示的形式公开。

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

Individual Participant Data (IPD) will not be shared, but statistical results will be published, reported, or presented.

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:

所有参与医生的答卷均通过“问卷星”这一在线调查平台收集。数据收集完成后,所有数据将从该在线平台永久删除,并安全保存至本地一台访问权限受限的设备中。

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

Responses from all participating physicians were collected via Questionnaire Star, an online survey platform. After data collection is complete, all data will be permanently deleted from the online platform and securely stored on a local, access-restricted device.

数据与安全监察委员会:

Data and Safety Monitoring Committee:

无/No

注册人:

Name of Registration:

 2025-12-29 15:41:21