ChiCTR2600120111 版本V1.0 版本创建时间2026/03/09 17:49:52 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:

ChiCTR2600120111 

最近更新日期:

Date of Last Refreshed on:

2026-03-09 17:49:44 

注册时间:

Date of Registration:

2026-03-09 00:00:00 

注册号状态:

预注册

Registration Status:

Prospective registration

注册题目:

利用人工智能综合分析技术预测前列腺癌的研究

Public title:

A Study on Predicting Prostate Cancer Using Artificial Intelligence-Based Integrated Analytical Techniques

注册题目简写:

English Acronym:

研究课题的正式科学名称:

基于人工智能多模态数据融合模型预测PSA灰区前列腺癌的临床应用研究

Scientific title:

Clinical Application Study of an Artificial Intelligence-Based Multimodal Data Fusion Model for Predicting Prostate Cancer in the PSA Gray Zone

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:

王晨 

研究负责人:

王晨 

Applicant:

Chen Wang 

Study leader:

Chen Wang 

申请注册联系人电话:

Applicant telephone:

+86 19817651720

研究负责人电话:

Study leader's
telephone:

+86 10 1234 5678

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:

wangchen@wmu.edu.cn

研究负责人电子邮件:

Study leader's E-mail:

782057771@qq.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:

浙江省丽水市括苍路289号

研究负责人通讯地址:

浙江省丽水市括苍路289号

Applicant address:

No. 289, KuaCang Road, Lishui City, Zhejiang Province

Study leader's address:

No. 289, KuaCang Road, Lishui City, Zhejiang Province

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:

温州医科大学第五附属医院(丽水市中心医院)

Applicant's institution:

The Fifth Affiliated Hospital of Wenzhou Medical College and Lishui Municipal Central Hospital

研究负责人所在单位:

丽水市中心医院

Affiliation of the Leader:

Lishui Central Hospital

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:

科研伦审2025(I)第333号(批)-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:

丽水市中心医院科研伦理审查委员会

Name of the ethic committee:

Scientific Research Ethics Committee of Lishui Central Hospital

伦理委员会批准日期:

Date of approved by ethic committee:

2025-10-28 00:00:00

伦理委员会联系人:

董丹妮

Contact Name of the ethic committee:

Dong DanNi

伦理委员会联系地址:

浙江省丽水市括苍路289号

Contact Address of the ethic committee:

No. 289, KuaCang Road, Lishui City, Zhejiang Province

伦理委员会联系人电话:

Contact phone of the ethic committee:

+86 578 2285719

伦理委员会联系人邮箱:

Contact email of the ethic committee:

16732020@qq.com

研究实施负责(组长)单位:

丽水市中心医院

Primary sponsor:

Lishui Central Hospital

研究实施负责(组长)单位地址:

浙江省丽水市括苍路289号

Primary sponsor's address:

No. 289, KuaCang Road, Lishui City, Zhejiang Province

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):

丽水市中心医院

具体地址:

浙江省丽水市括苍路289号

Institution
hospital:

Lishui Central Hospital

Address:

No. 289, KuaCang Road, Lishui City, Zhejiang Province

经费或物资来源:

2025年度省卫生健康行业科技计划项目

Source(s) of funding:

2025 Health Commission Science and Technology Project (Bureau-level)

研究疾病:

前列腺恶性肿瘤  

Target disease:

Prostate Malignant Tumor

研究疾病代码:

Target disease code:

研究类型:

诊断试验

Study type:

Diagnostic test

研究所处阶段:

其它 

Study phase:

N/A

研究设计:

诊断试验诊断准确性 

Study design:

Diagnostic test for accuracy 

研究目的:

本研究旨在通过整合 mpMRI 影像组学、血清非靶向代谢组学与重要临床指标,构建并验证一种可解释的多模态人工智能预测模型,用于提高 PSA 4–10 ng/mL(“灰区”)患者中前列腺癌的识别率与临床决策效率。具体目标为:提取并整合多模态特征,开发高性能的个体化风险预测模型;采用 SHAP 等可解释性方法构建模型解释框架,提升模型临床可解释性与医生采纳度;通过回顾性数据建模并在多中心前瞻性队列中验证模型的诊断性能与临床价值(包括提高穿刺阳性率与减少不必要穿刺的能力)。  

Objectives of Study:

This study aims to construct and validate an interpretable multimodal artificial intelligence prediction model by integrating mpMRI radiomics, serum untargeted metabolomics, and key clinical indicators, in order to improve the identification rate of prostate cancer and the efficiency of clinical decision-making in patients with PSA levels of 4–10 ng/mL (the 'gray zone'). The specific objectives are: to extract and integrate multimodal features and develop a high-performance individualized risk prediction model; to build a model explanation framework using interpretability methods such as SHAP, enhancing the model's clinical interpretability and physician acceptance; and to model retrospective data and validate the model's diagnostic performance and clinical value in a multicenter prospective cohort (including the ability to increase biopsy positivity rates and reduce unnecessary biopsies).

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 周岁; 2.PSA 4–10 ng/mL; 3.mpMRI 或超声提示可疑病灶; 4.既往未行前列腺穿刺活检,且既往无前列腺癌确诊史; 5.近 2 周内无急性尿路感染或导尿史; 6.同意采集血液样本并签署知情同意。

Inclusion criteria

1. Age ≥ 50 years old; 2. PSA 4–10 ng/mL; 3. mpMRI or ultrasound suggests suspicious lesions; 4. No previous prostate biopsy and no history of confirmed prostate cancer; 5. No acute urinary tract infection or catheterization in the past 2 weeks; 6. Agree to provide blood samples and sign informed consent.

排除标准:

1.既往诊断其他恶性肿瘤并在随访期内接受系统治疗者; 2.合并严重心血管、肝肾功能衰竭或精神障碍不能配合者; 3.长期(>3 个月)服用 5α-还原酶抑制剂者; 4.临床资料不全或关键影像/样本无法获得者; 5.MRI 图像质量差无法进行组学分析者; 6.撤回知情同意或研究者评估不宜继续者。

Exclusion criteria:

1. Patients with a history of other malignant tumors who received systemic treatment during the follow-up period; 2. Patients with severe cardiovascular, liver or kidney failure, or mental disorders who cannot cooperate; 3. Patients who have been taking 5α-reductase inhibitors for a long term (>3 months); 4. Patients with incomplete clinical data or key images/samples unavailable; 5. Patients whose MRI image quality is poor and cannot undergo radiomics analysis; 6. Patients who withdraw informed consent or are deemed unsuitable to continue by the researcher.

研究实施时间:

Study execute time:

From 2026-01-01 00:00:00 To 2028-12-31 00:00:00  

征募观察对象时间:

Recruiting time:

From 2026-03-10 00:00:00 To 2027-12-31 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):

本研究采用前列腺穿刺活检的病理结果作为诊断前列腺癌的金标准。

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):

Prostate biopsy pathology served as the gold standard for diagnosing prostate cancer in this study.

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):

基于人工智能多模态数据融合模型的前列腺癌风险预测系统。 该系统融合MRI影像组学特征、血清代谢组学特征及临床变量(如PSA、年龄、前列腺体积等),通过集成学习算法(SVM、随机森林、XGBoost、逻辑回归)构建预测模型,并结合SHAP可解释性分析模块生成个体化风险评估结果,用于辅助区分PSA灰区(4–10 ng/mL)患者的良恶性病变。

Index test:

Prostate cancer risk prediction system based on an artificial intelligence multimodal data fusion model. This system integrates MRI radiomic features, serum metabolomic features, and clinical variables (such as PSA, age, prostate volume, etc.), constructs a prediction model through ensemble learning algorithms (SVM, Random Forest, XGBoost, Logistic Regression), and combines an SHAP interpretability analysis module to generate individualized risk assessment results, which can assist in distinguishing benign and malignant lesions in patients within the PSA gray zone (4–10 ng/mL).

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)

本研究的目标人群为PSA灰区(4–10 ng/mL)前列腺癌筛查人群,包括在泌尿外科门诊或住院期间因PSA异常或影像提示可疑病灶而拟行穿刺活检的男性患者。

例数:

Sample size:

140

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):

The target population consisted of men aged ≥50 years in the PSA gray zone (4–10 ng/mL) with suspicious findings and no prior prostate cancer diagnosis who were scheduled for prostate biopsy.

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):

本研究的主要混淆疾病人群包括:良性前列腺增生(BPH);慢性前列腺炎;前列腺上皮内瘤变(PIN)或非典型腺体增生(ASAP)。

例数:

Sample size:

20

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

The main confounding conditions included benign prostatic hyperplasia (BPH), chronic prostatitis, and prostatic intraepithelial neoplasia (PIN) or atypical small acinar proliferation (ASAP), which can similarly elevate PSA and mimic imaging abnormalities in the PSA gray zone population.

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):

丽水市中心医院 

单位级别:

三级甲等 

Institution
hospital:

Lishui Central Hospital

Level of the institution:

Tertiary A

国家:

中国

省(直辖市):

浙江省 

市(区县):

 

Country:

China

Province:

Zhejiang

City:

单位(医院):

湖州市中心医院 

单位级别:

三级甲等 

Institution
hospital:

Huzhou Central Hospita

Level of the institution:

Tertiary A

国家:

中国

省(直辖市):

浙江省 

市(区县):

 

Country:

China

Province:

Zhejiang

City:

单位(医院):

宁波大学附属第一医院 

单位级别:

三级甲等 

Institution
hospital:

The First Affiliated Hospital of Ningbo University

Level of the institution:

Tertiary A

测量指标:

Outcomes:

指标中文名:

前列腺癌风险预测模型诊断准确性(AUC)

指标类型:

主要指标

Outcome:

Diagnostic accuracy of the prostate cancer risk prediction model (AUC)

Type:

Primary indicator

测量时间点:

患者完成MRI检查、血清样本采集及穿刺病理结果获得后(即确诊阶段),在数据清洗与标准化后统一进行模型预测与性能分析。

测量方法:

将患者的MRI影像组学特征、血清代谢组学特征及临床变量(包括PSA水平、年龄、前列腺体积等)输入至已训练并冻结的多模态人工智能模型,生成个体化风险预测结果;以穿刺活检病理结果为金标准,采用受试者工作特征曲线(ROC)分析计算模型曲线下面积(AUC),并结合灵敏度、特异度、阳性预测值、阴性预测值、准确率等指标评估模型诊断效能,使用DeLong检验比较不同模型AUC差异。

Measure time point of outcome:

After MRI, serum collection, and biopsy confirmation, data were standardized for model prediction an

Measure method:

MRI radiomic, serum metabolomic, and clinical variables were entered into a pre-trained multimodal AI model to generate individualized risk predictions, and diagnostic performance was evaluated against biopsy pathology using ROC analysis (AUC, sensitivity, specificity, PPV, NPV, accuracy) with DeLong testing for AUC comparisons.

采集人体标本:

Collecting sample(s)
from participants:

标本中文名:

外周血

组织:

Sample Name:

Blood

Tissue:

人体标本去向

使用后销毁  

说明

Fate of sample:

Destruction after use  

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 50 years
最大 Max age years

性别:

男性

Gender:

Male

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:

本研究采用标准化病例报告表(CRF)进行数据采集,并通过电子数据采集系统(EDC)进行数据录入与管理。所有数据在录入前进行双人核对,数据清洗后统一进行标准化处理。影像数据采用统一格式进行存储与脱敏处理,血清样本经规范流程采集与保存。研究数据由专人管理,确保数据安全性与完整性。

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

Data were collected using standardized case report forms (CRFs) and managed through an electronic data capture (EDC) system. Double data verification and data cleaning were performed before analysis. Imaging data were stored in a standardized and de-identified format, and serum samples were collected and processed according to standardized procedures. Data security and integrity were ensured by designated research personnel.

数据与安全监察委员会:

Data and Safety Monitoring Committee:

有/Yes

注册人:

Name of Registration:

 2026-03-09 17:49:44