ChiCTR2400091330 版本V1.0 版本创建时间2024/10/25 17:21:47 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:

ChiCTR2400091330 

最近更新日期:

Date of Last Refreshed on:

2024-10-25 17:21:42 

注册时间:

Date of Registration:

2024-10-25 00:00:00 

注册号状态:

预注册

Registration Status:

Prospective registration

注册题目:

大语言模型对胸部CT非结构化文本报告的辅助诊断价值

Public title:

The value of large language models to assist in the diagnosis of unstructured text reports of chest CT

注册题目简写:

大语言模型对胸部CT非结构化文本报告的辅助诊断价值

English Acronym:

The value of large language models to assist in the diagnosis of unstructured text reports of chest CT

研究课题的正式科学名称:

大语言模型对胸部CT非结构化文本报告的辅助诊断价值

Scientific title:

The value of large language models to assist in the diagnosis of unstructured text reports of chest CT

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:

范超凡 

研究负责人:

陈新 

Applicant:

Fan Chaofan 

Study leader:

Chen Xin 

申请注册联系人电话:

Applicant telephone:

+86 132 2940 4494

研究负责人电话:

Study leader's telephone:

+86 139 0220 5193

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:

1974799292@qq.com

研究负责人电子邮件:

Study leader's E-mail:

chen_xin1020@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:

广东省广州市海珠区工业大道中253号

研究负责人通讯地址:

广东省广州市海珠区工业大道中253号

Applicant address:

No. 253, Industrial Avenue Zhong, Haizhu District, Guangzhou, Guangdong, China

Study leader's address:

No. 253, Industrial Avenue Zhong, Haizhu District, Guangzhou, Guangdong, 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:

南方医科大学珠江医院

Applicant's institution:

Zhujiang Hospital, Southern Medical University

研究负责人所在单位:

南方医科大学珠江医院

Affiliation of the Leader:

Zhujiang Hospital, Southern Medical University

是否获伦理委员会批准:

是/Yes

Approved by ethic committee:

Yes

伦理委员会批件文号:

Approved No. of ethic committee:

2024-KY-302-01

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:

南方医科大学珠江医院医学伦理委员会

Name of the ethic committee:

Medical Ethics Committee, Zhujiang Hospital, Southern Medical University

伦理委员会批准日期:

Date of approved by ethic committee:

2024-09-30 00:00:00

伦理委员会联系人:

黄熙华

Contact Name of the ethic committee:

Huang Xihua

伦理委员会联系地址:

广东省广州市海珠区工业大道中253号

Contact Address of the ethic committee:

253 Gongye Avenue, Haizhu District, Guangzhou City, Guangdong Province

伦理委员会联系人电话:

Contact phone of the ethic committee:

+86 20 6278 3254

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:

南方医科大学珠江医院

Primary sponsor:

Zhujiang Hospital, Southern Medical University

研究实施负责(组长)单位地址:

广东省广州市海珠区工业大道中253号

Primary sponsor's address:

No. 253, Industrial Avenue Zhong, Haizhu District, Guangzhou, Guangdong, 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):

广东

市(区县):

广州

Country:

China

Province:

Guangdong

City:

Guangzhou

单位(医院):

南方医科大学珠江医院

具体地址:

广东省广州市海珠区工业大道中253号

Institution
hospital:

Zhujiang Hospital, Southern Medical University

Address:

253 Gongye Avenue, Haizhu District, Guangzhou City, Guangdong Province

经费或物资来源:

自筹

Source(s) of funding:

self-sponsor

Target disease:

Chest Diseases

Target disease code:

研究类型:

观察性研究

Study type:

Observational study

研究所处阶段:

回顾性研究 

Study phase:

Retrospective study

研究设计:

病例研究 

Study design:

Case study 

研究目的:

探讨当前主流的大语言模型(LLMs)包括GPT-4、GPT-3.5-turbo、Gemini-Pro和Qwen-max在胸部CT非结构化文本处理任务中的整体性能差异,并探讨有选项和开放式两种提问方式及微调策略对各类LLMs辅助胸部CT诊断性能的影响。  

Objectives of Study:

To explore the overall performance differences between the current mainstream large language models (LLMs) including GPT-4, GPT-3.5-turbo, Gemini-Pro, and Qwen-max in the unstructured text processing task of chest CT, and to explore the effects of two questioning modes, with options and open-ended, and the fine-tuning strategies on the performance of the various types of LLMs in assisting diagnosis in chest CT.

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)在院诊断包含慢性阻塞性肺疾病、肺部感染、肝囊肿、冠状动脉疾病、胆囊结石、肺心病、肺结核、肺气肿、脂肪肝、胸腔积液、支气管扩张、主动脉粥样硬化及肺大疱13种诊断中至少一种; (2)患者行胸部CT检查并有完整的胸部CT检查报告文本; (3)18-80岁。

Inclusion criteria

(1) The in-hospital diagnosis contained at least one of the 13 diagnoses of chronic obstructive pulmonary disease (COPD), pulmonary infection, liver cysts, coronary artery disease, gallbladder stones, pulmonary heart disease, tuberculosis, emphysema, fatty liver, pleural effusion, bronchiectasis, aortic coarctation, and pulmonary herpes; (2) The patient has undergone a chest CT examination and has a complete copy of the chest CT examination report. (3) 18-80 years old.

排除标准:

胸部CT检查报告文本不完整。

Exclusion criteria:

The text of the chest CT examination report was incomplete.

研究实施时间:

Study execute time:

From 2024-10-08 00:00:00 To 2025-08-28 00:00:00  

征募观察对象时间:

Recruiting time:

From 2024-11-01 00:00:00 To 2025-08-28 00:00:00  

干预措施:

Interventions:

组别:

有选项提问组

样本量:

7500

Group:

Multiple-choice questioning group

Sample size:

干预措施:

干预措施代码:

Intervention:

None

Intervention code:

组别:

无选项提问组

样本量:

7500

Group:

Open-ended questioning 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):

广东 

市(区县):

广州 

Country:

China 

Province:

Guangdong 

City:

Guangzhou 

单位(医院):

南方医科大学珠江医院 

单位级别:

三甲 

Institution
hospital:

Zhujiang Hospital, Southern Medical University

Level of the institution:

Tertiary A

测量指标:

Outcomes:

指标中文名:

主观回答准确率

指标类型:

主要指标

Outcome:

Subjective Answer Accuracy Rate

Type:

Primary indicator

测量时间点:

测量方法:

Measure time point of outcome:

Measure method:

指标中文名:

参考回答准确率

指标类型:

主要指标

Outcome:

Reference Answer Accuracy Rate

Type:

Primary indicator

测量时间点:

测量方法:

Measure time point of outcome:

Measure method:

指标中文名:

受试者工作特征曲线

指标类型:

次要指标

Outcome:

Receiver Operating Characteristic Curve

Type:

Secondary indicator

测量时间点:

测量方法:

Measure time point of outcome:

Measure method:

指标中文名:

受试者工作特征曲线下面积

指标类型:

次要指标

Outcome:

Area Under the Curve

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 18 years
最大 Max age 80 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

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):

2026年1月1日起,通过南方医科大学珠江医院临床研究中心平台公布,网址链接为https://zjyylcyjzx.wetrial.com/#。

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

Published on January 1, 2026 through the platform of clinical research center of Zhujiang Hospital of Southern Medical University , The URL link is https://zjyylcyjzx.wetrial.com/#.

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:

数据采集与管理将经过南方医科大学珠江医院Wetrial电子系统完成,网址链接为https://zjyylcyjzx.wetrial.com/#。

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

Data acquisition and managementwill becompleted by wetrial electronic systems of zhujiang hospital of southern medical university ,The URL link is https://zjyylcyjzx.wetrial.com/#.

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:

 2024-10-25 17:21:42