ChiCTR2400095031 版本V1.0 版本创建时间2024/12/31 16:28:08 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:

ChiCTR2400095031 

最近更新日期:

Date of Last Refreshed on:

2024-12-31 16:27:59 

注册时间:

Date of Registration:

2024-12-31 00:00:00 

注册号状态:

预注册

Registration Status:

Prospective registration

注册题目:

人工智能辅助淋巴结病理HE切片质量分级系统研究

Public title:

A?Study on?the Quality Grading System?of HE?Slices Assisted?by?Artificial Intelligence for Lymph Node Pathology

注册题目简写:

English Acronym:

研究课题的正式科学名称:

人工智能辅助淋巴结病理HE切片质量分级系统研究

Scientific title:

A?Study on?the Quality Grading System?of HE?Slices Assisted?by?Artificial Intelligence for Lymph Node Pathology

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:

罗清雅 

研究负责人:

罗清雅 

Applicant:

Luo qingya 

Study leader:

Luo qingya 

申请注册联系人电话:

Applicant telephone:

+86 13436189338

研究负责人电话:

Study leader's
telephone:

+86 23 68767645

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:

44084700@qq.com

研究负责人电子邮件:

Study leader's E-mail:

44084700@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:

重庆市沙坪坝区高滩岩西南医院教学楼14楼

研究负责人通讯地址:

重庆市沙坪坝区高滩岩正街29号

Applicant address:

No. 30, Gaotanyan Main Street, Shapingba District, Chongqing City

Study leader's address:

No 29 Gaotanyan Main Street Shapingba District Chongqing

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:

陆军军医大学第一附属医院

Applicant's institution:

The First Hospital Affiliated to Army Medical University

研究负责人所在单位:

中国人民解放军陆军军医大学第一附属医院

Affiliation of the Leader:

The First Affiliated Hospital of Army Medical University

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:

(B)KY2024321

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:

中国人民解放军陆军军医大学第一附属医院伦理委员会

Name of the ethic committee:

Ethics Committee of the First Affiliated Hospital of Army Medical University PLA

伦理委员会批准日期:

Date of approved by ethic committee:

2024-12-19 00:00:00

伦理委员会联系人:

贺莉

Contact Name of the ethic committee:

He Li

伦理委员会联系地址:

重庆市沙坪坝区高滩岩正街29号

Contact Address of the ethic committee:

No 29 Gaotanyan Main Street Shapingba District Chongqing

伦理委员会联系人电话:

Contact phone of the ethic committee:

+86 23 68754035

伦理委员会联系人邮箱:

Contact email of the ethic committee:

cqhl13@qq.com

研究实施负责(组长)单位:

中国人民解放军陆军军医大学第一附属医院

Primary sponsor:

The First Affiliated Hospital of Army Medical University

研究实施负责(组长)单位地址:

重庆市沙坪坝区高滩岩正街29号

Primary sponsor's address:

No 29 Gaotanyan Main Street Shapingba District Chongqing

试验主办单位(项目批准或申办者):

Secondary sponsor:

国家:

中国

省(直辖市):

重庆

市(区县):

Country:

China

Province:

Chongqing

City:

单位(医院):

中国人民解放军陆军军医大学第一附属医院

具体地址:

重庆市沙坪坝区高滩岩正街29号

Institution
hospital:

The First Affiliated Hospital of Army Medical University

Address:

No 29 Gaotanyan Main Street Shapingba District Chongqing

经费或物资来源:

2024年度临床研究病理诊断专项

Source(s) of funding:

The Special Project of Pathological Diagnosis in Clinical Research in 2024

研究疾病:

淋巴结HE切片质量控制  

Target disease:

quality control of HE slices in lymph nodes

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:

本研究拟通过基于深度学习人工智能在淋巴结HE切片质量控制评价中的辅助作用及其应用研究,实现对淋巴结切片质量进行准确分级,为常规病理质控提供一个确实、可靠、高效的评价系统。进而提高淋巴结HE切片质量,为病理医生能够准确进行诊断打下基础,并后期为技术人员操作问题个性化分析等方面提供数据支撑。  

Objectives of Study:

This study intends to achieve accurate grading of lymph node section quality and provide a reliable, reliable and efficient evaluation system for routine pathological quality control through the auxiliary role of deep learning-based artificial intelligence in the evaluation of lymph node section quality control and its application. It can improve the quality of lymph node HE section, lay a foundation for pathologists to make accurate diagnosis, and provide data support for technicians in the later stage of personalized analysis of operation problems.

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,性别不限;
2.穿刺及活检淋巴结病例;
3.镜下均是淋巴结组织;
4.可为同一患者不同淋巴结;

Inclusion criteria

1.There are no restrictions on age and genderCases of puncture and biopsy of lymph nodes;
2.Cases of puncture and biopsy of lymph nodes;
3.All cases were lymph node;
4.tissuesIt can be different lymph nodes in the same patient;

排除标准:

1.经镜下复核为非淋巴结组织;
2.穿刺组织长度<1cm;

Exclusion criteria:

1.The re-examination under the microscope confirmed that it was not lymph node tissue;
2.The length of the punctured tissue was <1cm;

研究实施时间:

Study execute time:

From 2025-01-01 00:00:00 To 2027-12-31 00:00:00  

征募观察对象时间:

Recruiting time:

From 2025-01-10 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):

3名高年资病理诊断医生,参考HE切片质量评价表进行评价,意见一致或大于2/3的评价意见为金标准。

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):

Three senior pathologists evaluated the quality of HE sections according to the evaluation table, and the gold standard was when the consensus or more than 2/3 of the evaluation opinions were consistent.

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):

本研究采用RBG颜色通道与HSV通道结合的双Ostu阈值分割算法去除病理切片中背景信息干扰。主要包括训练和验证两个步骤:模型训练时先在ImageNet数据集中预训练,作为切片质控分级模型的初始化参数,深度学习框架使用Pytorch,采用Swin_Transformer网络对病理切片进行深度学习。在模型测试过程中,病理切片经过Ostu算法去除背景信息后,输入模型对模型进行验证。其中,训练集和测试集按照3:1的比例进行划分。

Index test:

This study adopts a dual Ostu threshold segmentation algorithm integrating the RBG color channel and the HSV channel to eliminate the interference of background information in pathological sections. It mainly encompasses two steps: training and validation. In the process of model training, pre-train

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)

手术标本或穿刺标本送检淋巴结活检患者,并符合前述纳入排除标准的患者

例数:

Sample size:

600

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 whose surgical specimens or puncture specimens were submitted for lymph node biopsy and who met the aforementioned inclusion and exclusion criteria

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):

送检标本名称为淋巴结,镜下非淋巴组织

例数:

Sample size:

0

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

The name of the specimen was lymph node, and the microscopic non-lymphoid tissue

研究实施地点:

Countries of recruitment and research settings:

国家:

中国

省(直辖市):

重庆 

市(区县):

 

Country:

China

Province:

Chongqing

City:

单位(医院):

中国人民解放军陆军军医大学第一附属医院 

单位级别:

三级甲等 

Institution
hospital:

The First Affiliated Hospital of Army Medical University

Level of the institution:

Tertiary A

国家:

中国

省(直辖市):

重庆 

市(区县):

 

Country:

China

Province:

Chongqing

City:

单位(医院):

中国人民解放军陆军特色医学中心 

单位级别:

三级甲等 

Institution
hospital:

Army Medical Center of PLA

Level of the institution:

Tertiary A

国家:

中国

省(直辖市):

重庆 

市(区县):

 

Country:

China

Province:

Chongqing

City:

单位(医院):

中国人民解放军陆军军医大学第二附属医院 

单位级别:

三级甲等 

Institution
hospital:

The Second Affiliated Hospital of the Army Medical University

Level of the institution:

Tertiary A

测量指标:

Outcomes:

指标中文名:

真阳性率

指标类型:

主要指标

Outcome:

true positive rate

Type:

Primary indicator

测量时间点:

模型构建完成,深度学习数据后

测量方法:

将机器学习输出诊断结果与金标准进行对比,通过下列公式,计算敏感度,即真阳性率 真阳性率=真阳性人数÷金标准阳性人数

Measure time point of outcome:

After the model was constructed, the data were deeply learned

Measure method:

Compare the diagnosis results output by machine learning with the gold standard, and calculate the sensitivity, that is, the true positive, rate, throuah the following formula. True positive rate = true positive number/standard positive number.

指标中文名:

受试者工作曲线和曲线下面积

指标类型:

次要指标

Outcome:

ROC AUC

Type:

Secondary indicator

测量时间点:

模型构建完成,深度学习数据后

测量方法:

利用R语言或SPSS统计软件,绘制受试者工作曲线,计算约登指数

Measure time point of outcome:

After the model was constructed, the data were deeply learned

Measure method:

Using R language or SPsS statistical software, draw the working curve of the subjects and calculate the Jordan index.

指标中文名:

诊断速度

指标类型:

次要指标

Outcome:

diagnostic speed

Type:

Secondary indicator

测量时间点:

模型构建完成,深度学习数据后

测量方法:

病理切片图像数字化

Measure time point of outcome:

After the model was constructed, the data were deeply learned

Measure method:

Digitalization of pathological slides.

指标中文名:

漏诊率(假阴性率)

指标类型:

次要指标

Outcome:

false negative rate

Type:

Secondary indicator

测量时间点:

模型构建完成,深度学习数据后

测量方法:

将机器学习输出诊断结果与金标准进行对比,通过下列公式,计算漏诊率,即假阴性率 假阴性率=假阴性人数÷金标准阳性人数

Measure time point of outcome:

After the model was constructed, the data were deeply learned

Measure method:

Compare the diagnosis results output by machine learning with the gold standard, and calculate themissed diagnosis rate, that is, the false negative rate, through the following formula. False negative rate = number of false negative people/number of standard positive people.

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 0 years
最大 Max age 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):

不共享

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

Not to share

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:

不共享

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

Not to share

数据与安全监察委员会:

Data and Safety Monitoring Committee:

无/No

注册人:

Name of Registration:

 2024-12-31 16:27:59