|
注册号: Registration number: |
ChiCTR2600126597 |
|
最近更新日期: Date of Last Refreshed on: |
2026-06-11 18:05:13 |
|
注册时间: Date of Registration: |
2026-06-11 00:00:00 |
|
注册号状态: |
预注册 |
|
Registration Status: |
Prospective registration |
|
注册题目: |
基于自主代理的推理型智能体在消化道疾病诊疗中的应用:一项多中心随机对照试验 |
|
Public title: |
An Autonomous Agent-Based Reasoning Agents in the Diagnosis and Treatment of Digestive Diseases: A Multicenter Randomized Controlled Trial |
|
注册题目简写: |
|
|
English Acronym: |
|
|
研究课题的正式科学名称: |
基于自主代理的推理型智能体在消化道疾病诊 疗中的应用:一项多中心随机对照试验 |
|
Scientific title: |
An Autonomous Agent-Based Reasoning Agents in the Diagnosis and Treatment of Digestive Diseases: A Multicenter Randomized Controlled Trial |
|
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: |
田小朋 |
研究负责人: |
苏宁 |
|
Applicant: |
Xiaopeng Tian |
Study leader: |
Ning Su |
|
申请注册联系人电话: Applicant telephone: |
+86 138 0299 0523 |
研究负责人电话:
Study leader's |
+86 131 4571 0551 |
|
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: |
tianxp@sysucc.org.cn |
研究负责人电子邮件: Study leader's E-mail: |
snxkyy@gzhmu.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: |
广东省广州市越秀区横枝岗路62号 |
研究负责人通讯地址: |
广东省广州市越秀区横枝岗路62号 |
|
Applicant address: |
No. 62, Hengzhigang Road, Yuexiu District, Guangzhou City, Guangdong Province, China |
Study leader's address: |
No. 62, Hengzhigang Road, Yuexiu District, Guangzhou City, Guangdong Province, China |
|
申请注册联系人邮政编码: Applicant postcode: |
510060 |
研究负责人邮政编码: Study leader's postcode: |
510095 |
|
申请人所在单位: |
中山大学肿瘤防治中心 |
||
|
Applicant's institution: |
Sun Yat-sen University Cancer Center. |
||
|
研究负责人所在单位: |
广州市胸科医院 |
||
|
Affiliation of the Leader: |
Guangzhou Chest Hospital. |
||
|
是否获伦理委员会批准: |
是 |
||
|
Approved by ethic committee: |
Yes |
||
|
伦理委员会批件文号: Approved No. of ethic committee: |
IIT-2026-009 |
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: |
广州市胸科医院伦理委员会 |
||
|
Name of the ethic committee: |
Ethics Committee of Guangzhou Chest Hospital |
||
|
伦理委员会批准日期: Date of approved by ethic committee: |
2026-06-08 00:00:00 | ||
|
伦理委员会联系人: |
陶岚 |
||
|
Contact Name of the ethic committee: |
Lan Tao |
||
|
伦理委员会联系地址: |
广东省广州市越秀区横枝岗路62号 |
||
|
Contact Address of the ethic committee: |
No. 62, Hengzhigang Road, Yuexiu District, Guangzhou City, Guangdong Province, China |
||
|
伦理委员会联系人电话: Contact phone of the ethic committee: |
+86 20 8359 0406 |
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: |
广州市胸科医院 |
||||||||||||||||||||||
|
Primary sponsor: |
Guangzhou Chest Hospital |
||||||||||||||||||||||
|
研究实施负责(组长)单位地址: |
广东省广州市越秀区横枝岗路62号 |
||||||||||||||||||||||
|
Primary sponsor's address: |
No. 62, Hengzhigang Road, Yuexiu District, Guangzhou City, Guangdong Province, China |
||||||||||||||||||||||
|
试验主办单位(项目批准或申办者): Secondary sponsor: |
|
||||||||||||||||||||||
|
经费或物资来源: |
自筹,无额外经费来源 |
||||||||||||||||||||||
|
Source(s) of funding: |
Self-funded, with no additional sources of funding. |
||||||||||||||||||||||
|
研究疾病: |
消化道疾病 |
||||||||||||||||||||||
|
Target disease: |
Digestive diseases |
||||||||||||||||||||||
|
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: |
探索性研究/预试验 | ||||||||||||||||||||||
|
Study phase: |
0 |
||||||||||||||||||||||
|
研究设计: |
随机平行对照 |
||||||||||||||||||||||
|
Study design: |
Parallel |
||||||||||||||||||||||
|
研究目的: |
1. 主要目的:评估AI agent辅助医师决策相较于常规医师独立决策,在消化道疾病诊疗中能否提高诊断和治疗方案制定的正确率。 2. 次要目的: (1)比较两组在诊断领域的总体准确性,包括Top-3/Top-5诊断召回率、多标签诊断性能(F1分数)及诊断排序质量(nDCG)。 (2)比较两组在治疗领域的总体合理性,包括治疗推荐正确率、Top-3/Top-5治疗召回率及多方案治疗性能(F1分数)。 (3)评估两组在诊断和治疗层面与专家评审结论的一致性(Gwet's AC1系数)。 (4)比较两组的安全性表现,包括不安全推荐发生率、重大错误发生率、漏诊关键诊断率及禁忌治疗推荐率。 (5)比较两组的诊疗效率,包括达到最终管理决策所需时间、不必要检查率、适宜检查推荐率、单病例预估成本及与标准诊疗的成本一致性。 (6)比较两组的整体临床效用,包括诊断-治疗联合正确率、安全且正确率及专家总体偏好度。 |
||||||||||||||||||||||
|
Objectives of Study: |
1. Primary Objective To evaluate whether AI agent-assisted physician decision-making, compared with routine physician‑alone decision-making, improves the accuracy of diagnosis and treatment planning in the management of digestive tract diseases. 2. Secondary Objectives (1) To compare the overall diagnostic accuracy between the two groups, including top‑3/top‑5 diagnostic recall, multi‑label diagnostic performance (F1 score), and diagnostic ranking quality (nDCG). (2) To compare the overall therapeutic appropriateness between the two groups, including the correctness of treatment recommendations, top‑3/top‑5 treatment recall, and multi‑option treatment performance (F1 score). (3) To assess the agreement of diagnostic and therapeutic decisions with expert panel review between the two groups (Gwet's AC1 coefficient). (4) To compare safety outcomes between the two groups, including the incidence of unsafe recommendations, major errors, missed critical diagnoses, and contraindicated treatment recommendations. (5) To compare diagnostic and treatment efficiency between the two groups, including time to final management decision, rate of unnecessary testing, rate of appropriate test recommendations, estimated cost per patient, and cost consistency with standard care. (6) To compare the overall clinical utility between the two groups, including the combined correct rate for diagnosis‑treatment pairs, the safe‑and‑correct rate, and overall expert preference. |
||||||||||||||||||||||
|
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: |
1. 临床资料严重不完整,无法形成基本诊疗判断,或缺乏支持研究评估所需的关键信息; 2. 本次就诊未形成明确的诊疗决策或处理意见,无法追溯后续临床管理结果; 3. 研究期间退出研究,或因转诊至外院等原因导致主要结局信息无法获得; 4. 仅为健康体检、常规筛查或单纯咨询就诊,未进入实际诊断评估与诊疗决策流程者; 5. 研究者认为其他不适合纳入本研究的情况。 |
||||||||||||||||||||||
|
Exclusion criteria: |
1. Severely incomplete clinical data that prevents forming a basic diagnostic judgment, or lack of key information required to support study evaluations; 2. No clear diagnostic or treatment decision was made during the current visit, making it impossible to trace subsequent clinical management outcomes; 3. Withdrawal from the study during the research period, or inability to obtain primary outcome information due to transfer to another hospital or other reasons; 4. Visit was solely for health check-up, routine screening, or simple consultation, without entering the actual diagnostic evaluation and treatment decision-making process; 5. Other conditions deemed by the investigator as unsuitable for inclusion in this study. |
||||||||||||||||||||||
|
研究实施时间: Study execute time: |
从 From 2026-06-09 00:00:00至 To 2026-06-19 00:00:00 |
征募观察对象时间: Recruiting time: |
从 From 2026-06-13 00:00:00 至 To 2026-06-18 00:00:00 |
|
干预措施: Interventions: |
|
|
研究实施地点: Countries of recruitment and research settings: |
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
测量指标: Outcomes: |
|
|
采集人体标本:
Collecting sample(s)
|
|
|
征募研究对象情况: Recruiting status: |
尚未开始 Not yet recruiting |
年龄范围: Participant age: |
|
||||||
|
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): |
本研究采用多中心随机对照设计,在多家参与中心的消化内科门诊同步开展。符合入组条件的受试者按照1:1比例随机分配至常规医师决策组和AI agent辅助医师决策组。随机分组采用中心随机系统完成。 |
||||||||
|
Randomization Procedure (please state who generates the random number sequence and by what method): |
This study adopts a multicenter randomized controlled design, conducted simultaneously in the gastroenterology outpatient clinics of multiple participating centers. Eligible subjects are randomly assigned in a 1:1 ratio to either the conventional physician decision-making group or the AI agent-assisted physician decision-making group. The randomization is performed using a central randomization system. |
||||||||
|
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: |
结局评价由两名及以上专家在盲法条件下独立完成,对最终评估的独立专家为盲法。 |
|
Blinding: |
Outcome evaluation will be performed independently by two or more experts under blinded conditions, and the independent experts conducting the final assessment will also be blinded. |
|
试验完成后的统计结果(上传文件): |
|
|
Calculated Results after
|
|
|
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): |
数据存放平台为国家生物信息中心(https://ngdc.cncb.ac.cn/gsub/ ),原始数据无共享计划,如有需要,请通过电子邮件联系研究负责人以获取相关信息 |
|
The way of sharing IPD”(include metadata and protocol, If use web-based public database, please provide the url): |
The data is stored at the China National Center for Bioinformation (https://ngdc.cncb.ac.cn/gsub/ ). There is no plan to share the raw data. If needed, please contact the study leader via email for relevant information. |
|
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: |
数据采集根据病例记录表在各分中心进行,数据收集完毕后,由主要中心的采集和管理系统进行数据的统一管理,存储至专门的服务器中,由专人负责。 |
|
Data collection and Management (A standard data collection and management system include a CRF and an electronic data capture: |
Data collection will be carried out at each sub-center according to the case record forms. After data collection is completed, the data will be uniformly managed by the data collection and management system of the main center, stored on a dedicated server, and placed under the responsibility of designated personnel. |
|
数据与安全监察委员会: Data and Safety Monitoring Committee: |
有/Yes |