ChiCTR2600125476 版本V1.0 版本创建时间2026/05/27 11:56:12 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:

ChiCTR2600125476 

最近更新日期:

Date of Last Refreshed on:

2026-05-27 11:56:05 

注册时间:

Date of Registration:

2026-05-27 00:00:00 

注册号状态:

预注册

Registration Status:

Prospective registration

注册题目:

“睿麻助手”麻醉术前评估AI系统构建与临床验证

Public title:

Construction and Clinical Validation of "Ruima Assistant," an AI System for Preanesthesia Assessment

注册题目简写:

English Acronym:

研究课题的正式科学名称:

“睿麻助手”麻醉术前评估AI系统构建与临床验证

Scientific title:

Construction and Clinical Validation of "Ruima Assistant," an AI System for Preanesthesia Assessment

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:

陈潮金 

研究负责人:

陈潮金 

Applicant:

Chaojin Chen 

Study leader:

Chaojin Chen 

申请注册联系人电话:

Applicant telephone:

+86 134 3032 2182

研究负责人电话:

Study leader's
telephone:

+86 134 3032 2182

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:

chenchj28@mail.sysu.edu.cn

研究负责人电子邮件:

Study leader's E-mail:

chenchj28@mail.sysu.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:

广东省广州市天河区天河路600号

研究负责人通讯地址:

广东省广州市天河区天河路600号

Applicant address:

No. 600 Tianhe Road, Guangzhou, Guangdong Province

Study leader's address:

No. 600 Tianhe Road, Guangzhou, Guangdong 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:

中山大学附属第三医院

Applicant's institution:

The Third Affiliated Hospital of Sun Yat-sen University

研究负责人所在单位:

中山大学附属第三医院

Affiliation of the Leader:

The Third Affiliated Hospital of Sun Yat-sen 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:

中大附三医伦 II2026-164-02

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:

中山大学附属第三医院医学伦理委员

Name of the ethic committee:

Medical Ethics Committee of the Third Affiliated Hospital of Sun Yat-sen University

伦理委员会批准日期:

Date of approved by ethic committee:

2026-05-07 00:00:00

伦理委员会联系人:

黄凯琪

Contact Name of the ethic committee:

Kaiqi Huang

伦理委员会联系地址:

广东省广州市天河区天河路600号

Contact Address of the ethic committee:

No. 600 Tianhe Road, Guangzhou, Guangdong Province

伦理委员会联系人电话:

Contact phone of the ethic committee:

+86 85253302

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:

中山大学附属第三医院

Primary sponsor:

The Third Affiliated Hospital of Sun Yat-sen University

研究实施负责(组长)单位地址:

广东省广州市天河区天河路600号

Primary sponsor's address:

No. 600 Tianhe Road, Guangzhou, Guangdong 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):

广东省

市(区县):

广州市

Country:

China

Province:

Guangdong

City:

Guangzhou

单位(医院):

中山大学附属第三医院

具体地址:

广东省广州市天河区天河路600号

Institution
hospital:

The Third Affiliated Hospital of Sun Yat-sen University

Address:

No. 600 Tianhe Road, Guangzhou, Guangdong Province

经费或物资来源:

自筹经费

Source(s) of funding:

Self-funded

研究疾病:

无  

Target disease:

None

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:

其它 

Study phase:

N/A

研究设计:

随机平行对照 

Study design:

Parallel 

研究目的:

本研究旨在系统评价"睿麻助手"麻醉术前评估人工智能系统的构建效果、真实世界临床验证表现及其对青年麻醉医生术前评估能力的影响。 主要目的包括:(1)评价"睿麻助手"在真实临床环境中对术前评估关键信息抽取、风险评分和麻醉计划生成的准确性、一致性、完整性、可追溯性和安全性;(2)通过前瞻性受试者内交叉随机对照试验,评价AI辅助是否能够缩短青年麻醉医生在床旁访视前完成病历审阅和初步术前麻醉评估所需时间;(3)通过前瞻性自身前后对照研究,评价AI辅助是否能够提高青年麻醉医生麻醉计划质量,并改善关键风险识别与信息完整性。 次要目的包括:(1)分析不同院区、不同手术类别、不同ASA-PS层级及不同年龄分层下系统性能的稳定性;(2)分析青年麻醉医生对AI输出的采纳、修改与拒绝模式,探索AI辅助术前评估的人机协同特点;(3)评价系统在实际工作流中的可用性与可接受性,为后续扩展研究提供依据。  

Objectives of Study:

This study aims to systematically evaluate the construction, real-world clinical validation performance, and impact of "RuiMA-Assistant", an artificial intelligence system for preoperative anesthesia assessment, on the assessment competence of junior anesthesiologists. Primary objectives are: 1. To evaluate the accuracy, consistency, completeness, traceability, and safety of "RuiMA-Assistant" in extracting key preoperative information, generating risk scores, and producing anesthesia plans in a real clinical environment; 2. To evaluate, in a prospective within-subject crossover randomized controlled trial, whether AI assistance can shorten the time required for junior anesthesiologists to complete chart review and initial preoperative anesthesia assessment before bedside visit; 3. To evaluate, in a prospective within-subject before-after study, whether AI assistance can improve the quality of anesthesia plans drafted by junior anesthesiologists and enhance their identification of key risks and completeness of information. Secondary objectives are: 1. To analyze the stability of system performance across campuses, surgical categories, ASA-PS strata, and age strata; 2. To characterize patterns of acceptance, modification, and rejection of AI outputs by junior anesthesiologists, and to explore human-AI collaboration in preoperative assessment; 3. To evaluate system usability and acceptability in real clinical workflows, providing evidence for subsequent expansion studies.

药物成份或治疗方案详述:

 

Description for medicine or protocol of treatment in detail:

 

纳入标准:

患者受试者: 1. 研究期间于本院两院区进入常规术前评估流程的择期手术患者; 2. 拟接受全身麻醉、椎管内麻醉、神经阻滞或其他需麻醉科术前评估的麻醉方式; 3. 术前电子病历资料可完整获取。 麻醉医生受试者: 1. 在本院麻醉手术中心从事临床工作的青年麻醉医生,原则上从事麻醉临床工作年限 <= 5 年; 2. 自愿参加研究并签署书面知情同意书。

Inclusion criteria

Patient participants: 1. Elective surgical patients entering the routine preoperative assessment workflow at either of the two campuses during the study period; 2. Scheduled to receive general anesthesia, neuraxial anesthesia, peripheral nerve block, or other anesthesia requiring preoperative assessment by the anesthesiology department; 3. Complete preoperative electronic medical record available. Anesthesiologist participants: 1. Junior anesthesiologists currently practicing in the anesthesia and surgery center of the institution, in principle with <= 5 years of clinical anesthesia experience; 2. Voluntary participation with written informed consent.

排除标准:

患者排除标准:排除急诊手术、仅局部浸润麻醉且不需麻醉科系统性术前评估者、电子病历中关键文书或关键字段严重缺失而无法建立有效参考标签者。 麻醉医生排除标准:排除不愿参加研究或中途退出者、研究期间轮转安排无法完成规定病例评估者,以及已深度参与本研究金标准建立而无法满足盲法要求者。

Exclusion criteria:

Patient exclusion criteria: Exclude those undergoing emergency surgery, those receiving only local infiltration anesthesia without the need for systematic preoperative evaluation by the anesthesiology department, and those with critical documents or key fields missing in the electronic medical record, making it impossible to establish effective reference labels. Anesthesiologist exclusion criteria: Exclude those unwilling to participate in the study or who withdraw midway, those unable to complete the required case evaluations due to rotation arrangements during the study, and those who have been deeply involved in establishing the gold standard of this study, making it impossible to meet blinding requirements.

研究实施时间:

Study execute time:

From 2026-06-04 00:00:00 To 2026-06-18 00:00:00  

征募观察对象时间:

Recruiting time:

From 2026-06-04 00:00:00 To 2026-06-18 00:00:00

干预措施:

Interventions:

组别:

第一部分 系统性能前瞻性验证组

样本量:

1000

Group:

Part 1 — System performance prospective validation arm

Sample size:

干预措施:

干预措施代码:

Intervention:

None

Intervention code:

组别:

第二部分 AI 辅助评估组

样本量:

120

Group:

Part 2 — AI-assisted assessment arm

Sample size:

干预措施:

AI 辅助评估

干预措施代码:

Intervention:

AI-assisted assessment

Intervention code:

组别:

第二部分 常规评估对照组

样本量:

120

Group:

Part 2 — Conventional assessment control arm

Sample size:

干预措施:

常规评估

干预措施代码:

Intervention:

Conventional assessment

Intervention code:

组别:

第三部分 自身前后对照组

样本量:

80

Group:

Part 3 — Within-subject before-after arm

Sample size:

干预措施:

从同一病例来源中随机抽取 80 例患者,由 24 名青年麻醉医生按独立执业年限 1:1:1 分层(<1 年、1–3 年、>3 年各 8 人)进行评估;每例病例由三个年资分组各 1 名医生独立评估,每位医生承担 10 例。每位医生在同一病例上先在完全不接触 AI 输出的条件下按统一模板独立撰写初始麻醉方案(Physician),经系统加盖时间戳并强制锁定后提交;随后系统呈现同一病例的 AI 单独方案(AI-alone),医生在此基础上完成 AI 辅助下的修订方案(Physician+AI)并再次提交。同一组 80 例患者数据用于医生自身前后配对比较(Physician vs Physician+AI),并与 AI-alone 方案进行盲法对比。

干预措施代码:

Intervention:

80 patients are randomly drawn from the same case pool and evaluated by 24 junior anesthesiologists stratified 1:1:1 by years of independent practice (n=8 each for <1 year, 1–3 years, and >3 years); each case is independently evaluated by one anesthesiologist from each seniority stratum, with each anesthesiologist evaluating 10 cases. For every case, the anesthesiologist first independently drafts the initial anesthesia plan (Physician) under a standardized template while fully blinded to any AI output, and the plan is automatically time-stamped and locked upon submission; the system then presents the AI-alone plan for the same case, and the anesthesiologist completes and submits a revised plan under AI assistance (Physician+AI). The same 80 patients are used for within-subject paired comparison (Physician vs Physician+AI) and for blinded comparison against the AI-alone plan.

Intervention code:

研究实施地点:

Countries of recruitment and research settings:

国家:

中国

省(直辖市):

广东省 

市(区县):

广州市 

Country:

China

Province:

Guangdong

City:

Guangzhou

单位(医院):

中山大学附属第三医院 

单位级别:

三甲 

Institution
hospital:

The Third Affiliated Hospital of Sun Yat-sen University

Level of the institution:

Tertiary A

测量指标:

Outcomes:

指标中文名:

信息抽取F1值

指标类型:

主要指标

Outcome:

Information extraction F1-score

Type:

Primary indicator

测量时间点:

测量方法:

以专家参考标准为对照,按预设字段计算precision、recall与F1

Measure time point of outcome:

Measure method:

Precision, recall and F1 calculated against the expert reference standard on pre-specified fields

指标中文名:

风险评分一致性

指标类型:

主要指标

Outcome:

Risk-score agreement

Type:

Primary indicator

测量时间点:

测量方法:

ASA-PS、aCCI、RCRI、ARISCAT等评分与专家参考标准的Cohen's κ及加权κ

Measure time point of outcome:

Measure method:

Cohen's κ and weighted κ between system scores (ASA-PS, aCCI, RCRI, ARISCAT) and expert reference

指标中文名:

AI生成麻醉计划专家盲法总评分

指标类型:

主要指标

Outcome:

Blinded expert total score for AI-generated anesthesia plan

Type:

Primary indicator

测量时间点:

测量方法:

独立专家组采用统一五维量表(医学准确性、完整性、临床可用性、逻辑性与可读性、安全性)盲法评分,5分制

Measure time point of outcome:

Measure method:

Blinded scoring by an independent expert panel on a standardized 5-dimension Likert scale (medical accuracy, completeness, clinical usability, logic/readability, safety)

指标中文名:

床旁访视前完成病历审阅与初步术前麻醉评估的时间

指标类型:

主要指标

Outcome:

Time to complete chart review and initial preoperative assessment before bedside visit

Type:

Primary indicator

测量时间点:

测量方法:

对分组保持盲法的现场研究人员人工计时,以分钟为单位

Measure time point of outcome:

Measure method:

Manual timing in minutes by an on-site investigator blinded to allocation

指标中文名:

AI介入前后麻醉计划总评分差值

指标类型:

主要指标

Outcome:

Within-subject difference in total anesthesia-plan score before vs after AI assistance

Type:

Primary indicator

测量时间点:

测量方法:

同一医生同一病例上Physician+AI方案总分与Physician方案总分之差,由盲法专家组按统一五维量表评分

Measure time point of outcome:

Measure method:

Difference between Physician+AI and Physician total scores on the same case for the same anesthesiologist, scored by a blinded expert panel on the standardized 5-dimension scale

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:

尚未开始

Not yet recruiting

年龄范围:

Participant age:

最小 Min age 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):

第二部在研究启动前,由不参与现场评估、知情同意获取与结局评定的独立统计人员使用 R 4.3 统计软件,以青年麻醉医生为分层因素,在每位医生内部对其 10 例病例进行随机置换,生成 AI 辅助评估(5 例)与常规评估(5 例)1:1 的分配序列,从而保证每位医生内部分配严格平衡。随机种子(set.seed)预先固定并归档备查,分配序列存储于受密码保护的电子文件中,由不参与评估的研究协调员在每例评估开始前向医生逐例揭示,实现分配隐藏。第一部分与第三部分不涉及随机化。

Randomization Procedure (please state who generates the random number sequence and by what method):

Part 2 :Prior to study initiation, an independent statistician — not involved in on-site assessment, informed-consent acquisition, or outcome adjudication — uses R software version 4.3 to generate, with the junior anesthesiologist as the stratifying factor, a random permutation of the 10 assigned cases for each anesthesiologist, producing a 1:1 allocation between AI-assisted assessment (5 cases) and conventional assessment (5 cases), thereby ensuring strict within-anesthesiologist balance. Part 1 and Part 3 do not involve randomization.

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:

None

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

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:

本研究数据采集与管理由病例报告表(CRF)和电子数据采集系统(EDC)组成。研究启动前依据研究方案设计标准化eCRF,按子研究分模块(人口学、术前评估数据要素、风险评分、专家盲法五维度评分、第二部分计时、第三部分前后方案配对、可用性问卷),每项均设逻辑校验与必填规则。建立项目专属数据库,支持在线录入、自动逻辑核查、数据修改痕迹保留、权限分级与定期备份。原始病历数据经医院数据湖自动抽取并去标识化后导入EDC,“睿麻助手”系统输出随时间戳归档,所有数据以唯一研究编码替代身份信息。质量管理包括研究者培训、eCRF试运行、关键字段双人录入、定期一致性与逻辑核查,以及由项目负责人和统计人员联合签字的数据锁定流程。

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

Data collection and management consist of Case Report Forms (CRF) and an Electronic Data Capture (EDC) system. Standardized eCRFs are designed prior to study initiation according to the protocol, with sub-study modules (demographics, preoperative data elements, risk scoring, blinded five-dimension expert ratings, Part-2 timing, Part-3 paired before/after plans, and the usability questionnaire), each governed by logical validation and mandatory-field rules. Providing a project-specific database with online entry, automatic logical checks, full audit trails, role-based access, and scheduled backups. Raw EMR data are automatically extracted and de-identified via the institutional data lake before import; outputs of "RuiMA-Assistant" are archived with time stamps; all data are de-identified using unique study codes. Quality management includes investigator training, eCRF pilot testing, double entry for critical fields, periodic consistency and logical checks, and a data-lock procedure jointly signed by the principal investigator and the study statistician.

数据与安全监察委员会:

Data and Safety Monitoring Committee:

暂未确定/Not yet

注册人:

Name of Registration:

 2026-05-27 11:56:05