基于医疗多智能体的协同诊断系统关键技术研发与临床应用

注册号:

Registration number:

ChiCTR2600126778 

最近更新日期:

Date of Last Refreshed on:

2026-06-15 17:51:03 

注册时间:

Date of Registration:

2026-06-15 00:00:00 

注册号状态:

预注册

Registration Status:

Prospective registration

注册题目:

基于医疗多智能体的协同诊断系统关键技术研发与临床应用

Public title:

Research on Key Technologies and Clinical Application of Collaborative Diagnosis System Based on Medical Multi-Agent

注册题目简写:

English Acronym:

研究课题的正式科学名称:

基于医疗多智能体的协同诊断系统关键技术研发与临床应用

Scientific title:

Research on Key Technologies and Clinical Application of Collaborative Diagnosis System Based on Medical Multi-Agent

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:

虞玲华 

研究负责人:

虞玲华 

Applicant:

Linghua Yu 

Study leader:

Linghua Yu 

申请注册联系人电话:

Applicant telephone:

+86 158 2570 5735

研究负责人电话:

Study leader's
telephone:

+86 573 8208 2937

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:

ylhyqh@hotmail.com

研究负责人电子邮件:

Study leader's E-mail:

ylhyqh@hotmail.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:

嘉兴市中环南路1882号

研究负责人通讯地址:

嘉兴市中环南路1882号

Applicant address:

1882 Zhong Huan Nan Lu, Jiaxing

Study leader's address:

1882 Zhong Huan Nan Lu, Jiaxing

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:

嘉兴市第一医院

Applicant's institution:

First Hospital of Jiaxing

研究负责人所在单位:

嘉兴市第一医院

Affiliation of the Leader:

The First Hospital Of Jiaxing

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-LP-700

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:

嘉兴市第一医院医学伦理委员会

Name of the ethic committee:

Medical Ethics Committee of the First Hospital of Jiaxing

伦理委员会批准日期:

Date of approved by ethic committee:

2025-09-05 00:00:00

伦理委员会联系人:

许文

Contact Name of the ethic committee:

Wen Xu

伦理委员会联系地址:

嘉兴市中环南路1882号

Contact Address of the ethic committee:

1882 Zhong Huan Nan Lu, Jiaxing

伦理委员会联系人电话:

Contact phone of the ethic committee:

+86 573 8997 6378

伦理委员会联系人邮箱:

Contact email of the ethic committee:

xwkikimi@163.com

研究实施负责(组长)单位:

嘉兴市第一医院

Primary sponsor:

The First Hospital Of Jiaxing

研究实施负责(组长)单位地址:

嘉兴市中环南路1882号

Primary sponsor's address:

1882 Zhong Huan Nan Lu, Jiaxing

试验主办单位(项目批准或申办者):

Secondary sponsor:

国家:

中国

省(直辖市):

浙江省

市(区县):

Country:

China

Province:

Zhejiang

City:

单位(医院):

嘉兴市第一医院

具体地址:

嘉兴市中环南路1882号

Institution
hospital:

The First Hospital Of Jiaxing

Address:

1882 Zhong Huan Nan Lu, Jiaxing

经费或物资来源:

自选课题(自筹)

Source(s) of funding:

Provincial Applied Basic Research Program Project

研究疾病:

非酒精性脂肪性肝病  

Target disease:

Non-alcoholic fatty liver disease

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:

病例对照研究 

Study design:

Case-Control study 

研究目的:

1. 开发医疗多智能体协同诊断核心系统 整合多智能体技术、深度学习与医疗知识图谱,构建覆盖多模态数据的专科智能体群与协同决策机制,实现三大核心功能: (1) 多模态数据标准化处理:支持 DICOM 影像、WSI 病理切片、JSON 基因报告、非结构化病历的自动清洗与语义对齐; (2) 专科智能体协同推理:通过 “黑板模型 + 动态加权共识算法”,实现影像、病理、病历、基因智能体的信息共享与冲突协商,输出统一诊断意见; (3) 诊断结果可解释:采用 Grad-CAM 热力图、逻辑推理路径可视化,为医生提供直观诊断依据。 2. 构建多学科协同诊断全流程管理平台 基于医学大模型与物联网技术,建立覆盖 “数据采集 - 协同诊断 - 远程会诊 - 随访优化” 的全流程平台,实现三大目标: (1) 多源数据联合监测:实时整合患者影像、病理、检验、用药等数据,构建动态患者画像,支持疾病进展趋势预判; (2) 个性化诊断与干预闭环:针对不同疾病(如脂肪肝)输出多学科诊断建议(如影像分期 + 病理分型 + 基因靶点推荐),并联动临床制定干预方案; (3) 隐私保护型远程协同:结合联邦学习与差分隐私技术,实现 “数据不动模型动”,支持三甲医院与基层医院的跨机构智能会诊,推动优质资源下沉。 3. 临床应用示范与效果验证 以脂肪肝、结直肠癌为首批示范病种,在医院开展临床试用,预期达成: (1) 诊断效率提升 50%:将传统 MDT 会诊时间从 3-7 天缩短至 2 小时内; (2) 诊断准确率提升 15%-20% (3) 医生工作负荷降低 30%:减少医生数据整理、报告撰写等重复性工作,聚焦核心诊断决策。  

Objectives of Study:

Develop a Core System for Collaborative Diagnosis with Medical Multi-Agent Systems Integrate multi-agent technology, deep learning, and medical knowledge graphs to construct specialized intelligent agent groups and collaborative decision-making mechanisms that cover multimodal data, achieving three core functions: 1. Standardized processing of multimodal data: support automatic cleaning and semantic alignment of DICOM images, WSI pathological sections, JSON genetic reports, and unstructured medical records; 2. Collaborative reasoning among specialized intelligent agents: realize information sharing and conflict negotiation among imaging, pathology, medical records, and genetic intelligent agents through the "blackboard model + dynamic weighted consensus algorithm", outputting unified diagnostic opinions; 3. Explainable diagnosis results: provide doctors with intuitive diagnostic evidence using Grad-CAM heatmaps and visualization of logical inference paths. Build a Full-Process Management Platform for Multidisciplinary Collaborative Diagnosis Establish a full-process platform covering "data collection - collaborative diagnosis - remote consultation - follow-up optimization" based on large-scale medical models and IoT technology, achieving three goals: 1. Joint monitoring of multi-source data: real-time integration of patient imaging, pathology, laboratory tests, medication, etc., to build dynamic patient profiles supporting disease progression prediction; 2. Personalized diagnosis and intervention closed-loop: output multidisciplinary diagnostic recommendations (e.g., imaging staging + pathological classification + gene target recommendations) for different diseases (e.g., fatty liver disease), and connect with clinical practice to formulate intervention plans; 3. Privacy-preserving remote collaboration: combine federated learning and differential privacy technology to achieve "data-stationary, model-mobile" cross-institutional intelligent consultation between tertiary hospitals and primary healthcare facilities, promoting the distribution of high-quality medical resources. Clinical Application Demonstration and Effectiveness Validation Taking fatty liver disease and colorectal cancer as the first demonstration diseases, conduct clinical trials in hospitals, with expected achievements: 1. Diagnosis efficiency increased by 50%: reduce traditional MDT consultation time from 3-7 days to within 2 hours; 2. Diagnosis accuracy increased by 15%-20%; 3. Physician workload reduced by 30%: reduce repetitive tasks such as data organization and report writing for doctors, allowing them to focus on core diagnostic decisions.

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.年龄小于16岁或大于90岁;

Exclusion criteria:

1.Less than 16 or greater than 90;

研究实施时间:

Study execute time:

From 2026-06-16 00:00:00 To 2028-12-31 00:00:00  

征募观察对象时间:

Recruiting time:

From 2026-06-17 00:00:00 To 2027-12-31 00:00:00

干预措施:

Interventions:

组别:

脂肪肝组

样本量:

1000

Group:

Fatty liver group

Sample size:

干预措施:

干预措施代码:

Intervention:

None

Intervention code:

组别:

健康对照组

样本量:

100

Group:

Healthy control 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):

浙江省 

市(区县):

 

Country:

China

Province:

Zhejiang

City:

单位(医院):

嘉兴市第一医院 

单位级别:

三级甲等 

Institution
hospital:

The First Hospital Of Jiaxing

Level of the institution:

Tertiary A

测量指标:

Outcomes:

指标中文名:

肝脏健康-代谢指标

指标类型:

主要指标

Outcome:

Liver Health - Metabolic Indicators

Type:

Primary indicator

测量时间点:

无确切时间点

测量方法:

基于医疗多智能体的协同,实现对患者多模态数据的深度解析

Measure time point of outcome:

None

Measure method:

Leveraging medical multi-agent collaboration to achieve deep analysis of patients' multimodal data.

指标中文名:

体重

指标类型:

次要指标

Outcome:

Weight

Type:

Secondary indicator

测量时间点:

无确切时间点

测量方法:

受试者空腹状态下脱去厚重衣物,站立于校准后的电子体重秤中央平稳静置,读取稳定数值完成体重测量。

Measure time point of outcome:

None

Measure method:

Subjects remove heavy clothing in a fasting state, stand steadily in the center of a calibrated electronic weight scale, and record stable readings to complete weight measurement.

采集人体标本:

Collecting sample(s)
from participants:

标本中文名:

组织:

Sample Name:

NA

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 16 years
最大 Max age 90 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:

不公开/Private

盲法:

Blinding:

None

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):

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

None

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

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

Case Record Form

数据与安全监察委员会:

Data and Safety Monitoring Committee:

无/No

注册人:

Name of Registration:

 2026-06-15 17:49:13