ChiCTR2600125133 版本V1.0 版本创建时间2026/05/21 15:44:02 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:

ChiCTR2600125133 

最近更新日期:

Date of Last Refreshed on:

2026-05-21 15:43:33 

注册时间:

Date of Registration:

2026-05-21 00:00:00 

注册号状态:

补注册

Registration Status:

Retrospective registration

注册题目:

阿尔茨海默病的智能影像辅助诊断及预测系统研发

Public title:

Research and Development of an Intelligent Imaging-Assisted Diagnosis and Prediction System for Alzheimer's Disease.

注册题目简写:

English Acronym:

研究课题的正式科学名称:

阿尔茨海默病的智能影像辅助诊断及预测系统研发

Scientific title:

Research and Development of an Intelligent Imaging-Assisted Diagnosis and Prediction System for Alzheimer's Disease.

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:

陈宗勤 

研究负责人:

李郁欣 

Applicant:

Chen Zongqin 

Study leader:

Li Yuxin 

申请注册联系人电话:

Applicant telephone:

+86 52887466

研究负责人电话:

Study leader's telephone:

+86 52887466

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:

19960646690@163.com

研究负责人电子邮件:

Study leader's E-mail:

liyuxin@fudan.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:

上海市乌鲁木齐中路12号

研究负责人通讯地址:

上海市乌鲁木齐中路12号

Applicant address:

No. 12, Urumqi Middle Road, Shanghai, China

Study leader's address:

No. 12, Urumqi Middle Road, Shanghai, China

申请注册联系人邮政编码:

Applicant postcode:

200040

研究负责人邮政编码:

Study leader's postcode:

200040

申请人所在单位:

复旦大学附属华山医院

Applicant's institution:

Huashan Hospital, Fudan University

研究负责人所在单位:

复旦大学附属华山医院

Affiliation of the Leader:

Huashan Hospital, Fudan 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:

(2025)临审第(1570)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:

复旦大学附属华山医院伦理审查委员会

Name of the ethic committee:

Institutional Review Board of Huashan Hospital, Fudan University

伦理委员会批准日期:

Date of approved by ethic committee:

2025-12-17 00:00:00

伦理委员会联系人:

吴翠云

Contact Name of the ethic committee:

Wu Cuiyun

伦理委员会联系地址:

上海市乌鲁木齐中路12号

Contact Address of the ethic committee:

No. 12, Urumqi Middle Road, Shanghai, China

伦理委员会联系人电话:

Contact phone of the ethic committee:

+86 21 5288 8045

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:

复旦大学附属华山医院

Primary sponsor:

Huashan Hospital, Fudan University

研究实施负责(组长)单位地址:

上海市乌鲁木齐中路12号

Primary sponsor's address:

No. 12, Urumqi Middle Road, Shanghai, 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):

上海

市(区县):

上海

Country:

China

Province:

Shanghai

City:

Shanghai

单位(医院):

复旦大学附属华山医院

具体地址:

上海市乌鲁木齐中路12号

Institution
hospital:

Huashan Hospital, Fudan University

Address:

No. 12, Urumqi Middle Road, Shanghai, China

经费或物资来源:

复旦大学附属华山医院

Source(s) of funding:

Huashan Hospital, Fudan University

Target disease:

Alzheimer's disease

Target disease code:

研究类型:

诊断试验

Study type:

Diagnostic test

研究所处阶段:

回顾性研究 

Study phase:

Retrospective study

研究设计:

诊断试验诊断准确性 

Study design:

Diagnostic test for accuracy 

研究目的:

本项目旨在开发并验证一套面向MCI/AD患者的“早期诊断-新药监测-精准预测”全流程智能辅助诊疗及预测系统,具体目的: 1.攻克早期诊断瓶颈:提出并构建“基于跨模态感知合成的智能诊断模型”,利用 MRI 合成多尺度 PET 影像,将跨模态影像的关联关系与先验脑图谱进行融合,生成具有脑区感知能力的高维表征,能够在仅有3D MRI数据的场景下,可视化合成缺失的PET影像,并且生成的脑区感知模块能够帮助实现AD的早期精准诊断,提升临床前或MCI的检出率。 2.突破新药疗效与安全性监测滞后性:利用“基于掩码自监督学习的新药不良反应检测模型”,提高对AD新药疗效动态观察中的潜在不良反应(主要为ARIA)的早期发现和监测能力,为临床用药安全性提供实时、智能化的辅助诊断工具,减少放射科医生在阅片过程因为肉眼计数与人工标注可能出现的漏诊,提高临床用药安全性。 3.实现个体化进展精准预测:引入“基于序约束损失函数的个体化预测模型”,充分捕捉疾病进展的非线性、异质性轨迹。该模型旨在为每位AD患者提供高度个性化的病情进展(如认知功能下降速度、向痴呆转化风险、功能丧失能)概率预测和风险分层,显著提升预测的准确性和临床实用性,辅助临床医生制定个体化干预和管理策略。 最终,整合上述三个核心创新模型,构建一个统一的、基于人工智能的全流程智能辅助诊疗及预测平台。该平台将利用收集的大规模回顾性多中心数据进行模型构建与训练,并致力于在真实世界临床环境中进行验证和应用,最终实现:①提升AD诊疗效率与精准度:为医生提供强大的辅助诊断、用药监测和预后判断工具;②优化患者管理:实现更早干预、更安全的的药物治疗和更精准的个体化照护计划;③推动AD研究与药物开发:为理解疾病异质性、评估新药疗效提供新的数据支持。  

Objectives of Study:

This project aims to develop and validate a comprehensive intelligent imaging-assisted diagnosis and treatment system for patients with MCI/AD, encompassing an "early diagnosis – new drug monitoring – accurate prediction" full-process platform. The specific objectives are as follows: 1. Address the bottleneck of early diagnosis: Propose and develop an "intelligent diagnostic model based on cross-modal perception and synthesis." This model utilizes MRI to synthesize multi-scale PET images, integrating the correlations between cross-modal images with prior brain atlases to generate high-dimensional representations with brain region awareness. In scenarios where only 3D MRI data are available, the model can visually synthesize missing PET images. The resulting brain region-aware module facilitates early and accurate diagnosis of AD, thereby improving the detection rate of preclinical AD or MCI. 2. Overcome the lag in monitoring the efficacy and safety of new drugs: Utilize a "masked self-supervised learning-based adverse drug reaction detection model" to enhance the early detection and monitoring of potential adverse reactions (primarily ARIA) during dynamic observation of new drug efficacy in AD. This provides clinicians with a real-time, intelligent decision-support tool for medication safety, reducing missed detections that may occur due to manual counting and annotation by radiologists, thus improving clinical medication safety. 3. Achieve accurate individualized progression prediction: Introduce an "ordinal constraint loss function-based individualized prediction model" to fully capture the nonlinear and heterogeneous trajectories of disease progression. This model aims to provide highly personalized probabilistic predictions and risk stratification for each AD patient, including the rate of cognitive decline, risk of conversion to dementia, and loss of function. It will significantly improve the accuracy and clinical utility of predictions, assisting clinicians in developing individualized intervention and management strategies. 4. Develop an integrated full-process intelligent platform: Ultimately, integrate the three core innovative models described above to build a unified, AI-based full-process intelligent assisted diagnosis and treatment platform. The platform will be constructed and trained using large-scale, retrospective, multi-center data, and will be validated and applied in real-world clinical settings. The ultimate goals are to: 1)Improve the efficiency and accuracy of AD diagnosis and treatment: Provide physicians with powerful tools for assisted diagnosis, medication monitoring, and prognosis assessment. 2)Optimize patient management: Enable earlier intervention, safer drug therapy, and more precise individualized care plans. 3)Advance AD research and drug development: Provide new data support for understanding disease heterogeneity and evaluating the efficacy of new drugs.

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.年龄50-85岁,性别不限; 2.基于NIA-AA临床诊断标准,被明确诊断为AD/MCI; 3.具有至少包含3D T1WI、SWI、Flair序列的MRI影像资料; 4.符合并接受DMT新药治疗,且有多次随诊影像资料。

Inclusion criteria

1. Aged 50-85 years, both genders eligible; 2. Confirmed diagnosis of AD/MCI according to NIA-AA clinical diagnostic criteria; 3. Availability of MRI data comprising at least 3D T1WI, SWI, and FLAIR sequences; 4. Eligible for and undergoing DMT novel drug treatment, with multiple follow-up imaging records available.

排除标准:

1.其他类型痴呆症患者及重大脑部疾病; 2.有严重(难治性)抑郁、焦虑、精神分裂症等精神类疾病,或有医学共存疾病影响生存期。

Exclusion criteria:

1. Patients with other types of dementia or major brain diseases; 2. Presence of severe (treatment-resistant) depression, anxiety, schizophrenia, or other psychiatric disorders, or medical comorbidities that may affect life expectancy.

研究实施时间:

Study execute time:

From 2025-08-01 00:00:00 To 2027-07-31 00:00:00  

征募观察对象时间:

Recruiting time:

From 2025-09-04 00:00:00 To 2026-07-31 00:00:00  

诊断试验:

Diagnostic Tests:

金标准或参考标准(即可准确诊断某疾病的单项方法或多项联合方法,在本研究中用于诊断是否有该病的临床参考标准):

临床专家基于MRI影像(SWI、Flair序列等)对ARIA-E(水肿)和ARIA-H(微出血)的联合判读结果,由双人“背靠背”阅片,不一致时由高年资专家仲裁确定。

Gold Standard or Reference Standard (The clinical reference standards required to establish the presence or absence of the target condition in the tested population in present study):

The combined interpretation results for ARIA-E (edema) and ARIA-H (microhemorrhage) by clinical experts based on MRI images (SWI, FLAIR sequences, etc.) were obtained through double "back-to-back" reading. Any disagreements were resolved by arbitration from a senior expert.

指标试验(即本研究的待评估诊断试验,无论为方法、生物标志物或设备,均请列出名称):

基于掩码自监督学习(MAE算法)的ARIA自动化检测模型,输入为3D T1WI、SWI、Flair序列,输出为ARIA有无及严重程度评估。

Index test:

An automated ARIA detection model based on masked self-supervised learning (MAE algorithm), with 3D T1WI, SWI, and FLAIR sequences as inputs, and the presence/absence of ARIA and severity assessment as outputs.

目标人群(可以是某种疾病患者或正常人群,详细描述其疾病特征,注意应纳入符合分布特点的全序列病例,具有良好的代表性)

明确诊断为阿尔茨海默病(AD)或轻度认知障碍(MCI)的患者,年龄50-85岁,接受疾病修饰治疗(DMT)新药治疗,具有多次随访MRI影像资料(含3D T1WI、SWI、Flair序列)。排除其他类型痴呆、重大脑部疾病及严重精神类疾病患者。

例数:

Sample size:

350

Target condition (The target condition is a particular disease or disease stage that the index test will be intended to identify. Please specify the characteristics in detail; the population should has a complete spectrum and good representative):

Patients definitively diagnosed with Alzheimer's disease (AD) or mild cognitive impairment (MCI), aged 50–85 years, receiving disease-modifying therapy (DMT) with novel drugs, and having multiple follow-up MRI scans (including 3D T1WI, SWI, and FLAIR sequences). Patients with other types of dementia, major brain diseases, or severe psychiatric disorders are excluded.

容易混淆的疾病人群(即与目标疾病不易区分的一种或多种不同疾病,应避免采用正常人群对照的病例-对照设计):

其他类型的痴呆症患者(如血管性痴呆、额颞叶痴呆、路易体痴呆等),以及伴有脑微出血或脑白质病变的非AD脑小血管疾病患者。

例数:

Sample size:

100

Population with condition difficult to distinguish from the target condition, the normal population in a case-control study design should be avoid:

Patients with other types of dementia (e.g., vascular dementia, frontotemporal dementia, dementia with Lewy bodies) and those with non-AD cerebral small vessel disease presenting with cerebral microbleeds or white matter lesions were excluded.

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):

浙江大学医学院附属第一医院 

单位级别:

三甲 

Institution
hospital:

The First Affiliated Hospital, Zhejiang University School of Medicine

Level of the institution:

Tertiary A

国家:

中国

省(直辖市):

上海市 

市(区县):

 

Country:

China 

Province:

Shanghai 

City:

 

单位(医院):

上海交通大学医学院附属同仁医院 

单位级别:

三甲 

Institution
hospital:

Tongren Hospital, Shanghai Jiao Tong University School of Medicine

Level of the institution:

Tertiary A

国家:

中国

省(直辖市):

福建省 

市(区县):

 

Country:

China 

Province:

Fujian 

City:

 

单位(医院):

福建医科大学附属协和医院 

单位级别:

三甲 

Institution
hospital:

Union Hospital, Fujian Medical University

Level of the institution:

Tertiary A

测量指标:

Outcomes:

指标中文名:

淀粉样蛋白相关影像学异常(ARIA)总体检出率

指标类型:

主要指标

Outcome:

Overall incidence of Amyloid-Related Imaging Abnormalities (ARIA)

Type:

Primary indicator

测量时间点:

基线、治疗后 3 个月、6 个月、12 个月及不良事件发生时

测量方法:

采用 3.0T MRI 的 SWI/FLAIR 序列,按照国际 ARIA 评估标准,由 2 名高年资影像科医师盲法评估,记录治疗期间任何时间点出现的 ARIA(包括 ARIA-E 和 ARIA-H)的检出率

Measure time point of outcome:

Baseline, 3 months, 6 months, 12 months after treatment, and at the time of adverse events.

Measure method:

ARIA (including ARIA-E and ARIA-H) were assessed using 3.0T MRI SWI/FLAIR sequences, evaluated by two blinded senior radiologists according to international ARIA criteria. The overall incidence of ARIA during treatment was recorded.

指标中文名:

灵敏度

指标类型:

主要指标

Outcome:

Sensitivity

Type:

Primary indicator

测量时间点:

测量方法:

Measure time point of outcome:

Measure method:

指标中文名:

特异度

指标类型:

主要指标

Outcome:

Specificity

Type:

Primary 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:

正在进行

Recruiting

年龄范围:

Participant age:

最小 Min age 50 years
最大 Max age 85 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:

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):

如需原始数据可电邮负责人共享

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

If you need raw data, you can share it with the person in charge via email

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:

ResMan、PACS

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

ResMan, PACS

数据与安全监察委员会:

Data and Safety Monitoring Committee:

有/Yes

注册人:

Name of Registration:

 2026-05-21 15:43:33