中国临床试验注册中心 - 世界卫生组织国际临床试验注册平台一级注册机构

注册号： Registration number：	ChiCTR2500111128
最近更新日期： Date of Last Refreshed on：	2025-10-27 10:04:10
注册时间： Date of Registration：	2025-10-27 00:00:00
注册号状态：	预注册
Registration Status：	Prospective registration
注册题目：	急性视网膜坏死综合征(ARN) AI 辅助诊疗及预后评估模型的构建与验证
Public title：	Construction and Validation of an AI-Assisted Diagnosis, Treatment, and Prognostic Evaluation Model for Acute Retinal Necrosis Syndrome (ARN)
注册题目简写：
English Acronym：
研究课题的正式科学名称：	急性视网膜坏死综合征(ARN) AI 辅助诊疗及预后评估模型的构建与验证
Scientific title：	Construction and Validation of an AI-Assisted Diagnosis, Treatment, and Prognostic Evaluation Model for Acute Retinal Necrosis Syndrome (ARN)
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：	王小雪	研究负责人：	张育
Applicant：	Wang Xiaoxue	Study leader：	Zhang Yu
申请注册联系人电话： Applicant telephone：	+86 137 5349 8916	研究负责人电话： Study leader's telephone：	+86 351 828 6884
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：	2145962552@qq.com	研究负责人电子邮件： Study leader's E-mail：	2145962552@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：	中国山西省太原市杏花岭区府东街100号	研究负责人通讯地址：	中国山西省太原市杏花岭区府东街100号
Applicant address：	100 Fudong Street, Xinghualing District, Taiyuan 030000, Shanxi, China	Study leader's address：	100 Fudong Street, Xinghualing District, Taiyuan 030000, Shanxi, China
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：	山西医科大学
Applicant's institution：	Shanxi Medical University
研究负责人所在单位：	山西省眼科医院
Affiliation of the Leader：	Shanxi Eye Hospital

是否获伦理委员会批准：	是
Approved by ethic committee：	Yes
伦理委员会批件文号： Approved No. of ethic committee：	SXYYLL-KSSC038	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：	山西省眼科医院医学伦理委员会
Name of the ethic committee：	Medical Ethics Committee of Shanxi Eye Hospital
伦理委员会批准日期： Date of approved by ethic committee：	2025-09-24 00:00:00
伦理委员会联系人：	高妍
Contact Name of the ethic committee：	Gao Yan
伦理委员会联系地址：	中国山西省太原市杏花岭区府东街100号
Contact Address of the ethic committee：	100 Fudong Street, Xinghualing District, Taiyuan 030000, Shanxi, China
伦理委员会联系人电话： Contact phone of the ethic committee：	+86 351 828 6884	伦理委员会联系人邮箱： Contact email of the ethic committee：	tri_cat@163.com

研究实施负责（组长）单位：

山西省眼科医院

Primary sponsor：

Shanxi Eye Hospital

研究实施负责（组长）单位地址：

中国山西省太原市杏花岭区府东街100号

Primary sponsor's address：

100 Fudong Street, Xinghualing District, Taiyuan 030000, Shanxi, 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)：	山西	市(区县)：	太原
Country：	China	Province：	Shanxi	City：	Taiyuan
单位(医院)：	山西省眼科医院	具体地址：	中国山西省太原市杏花岭区府东街100号
Institution hospital：	Shanxi Eye Hospital	Address：	100 Fudong Street, Xinghualing District, Taiyuan 030000, Shanxi, China

经费或物资来源：

自筹

Source(s) of funding：

Self-funded

研究疾病：

急性视网膜坏死综合征

Target disease：

Acute Retinal Necrosis Syndrome

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：

单臂

Study design：

Single arm

研究目的：

本课题聚焦急性视网膜坏死综合征(Acute Retinal Necrosis, ARN)诊疗中早期诊断难、治疗缺乏标准化及预后评估困难等核心问题。研究以本院2015年起收治的ARN病例队列为基础，系统构建了涵盖人口学基线、病原学病毒分型、视网膜血管闭塞范围、抗病毒治疗时机、手术干预方式选择，以及术后最佳矫正视力、视网膜脱离等并发症在内的23项核心观测指标，并据此建立了标准化临床数据库。本研究采用XGBoost算法，构建了两类关键预测模型：其一为治疗策略优化模型，旨在为抗病毒药物联合方案及玻璃体切除术介入时机提供循证决策支持；其二为预后风险评估模型，用于量化预测患者在12个月内发生视网膜脱离及视力下降至0.1以下的中期风险。研究引入SHAP(SHapley Additive exPlanations)值解析框架，以明确各影响因素对模型预测结果的贡献的贡献权重，从而提高模型决策透明度。本研究将通过纳入新增患者进行内部验证与模型迭代优化，最终形成适配临床实践的ARN精准诊疗辅助工具。该工具旨在通过个性化治疗推荐与预后预警，有效降低ARN所致视力丧失风险，推动诊疗流程的标准化与规范化。为基层医疗机构同类疾病的精准管理提供参考范式，具有显著的临床意义与应用价值。

Objectives of Study：

This research focuses on addressing the core challenges in the diagnosis and treatment of Acute Retinal Necrosis (ARN), including difficulties in early diagnosis, lack of standardization in treatment, and challenges in prognostic evaluation. Based on the cohort of ARN patients admitted to our hospital since 2015, the study systematically established 23 core observation indicators. These indicators cover demographic baselines, etiological virus typing, the scope of retinal vascular occlusion, the timing of antiviral treatment, the selection of surgical intervention methods, as well as postoperative outcomes such as best-corrected visual acuity and complications like retinal detachment. A standardized clinical database was constructed based on these indicators. This study employed the XGBoost algorithm to develop two key predictive models. The first is a treatment strategy optimization model, which aims to provide evidence-based decision support for the combination of antiviral drugs and the timing of vitrectomy intervention. The second is a prognostic risk assessment model, designed to quantitatively predict the medium-term risks of patients developing retinal detachment and visual acuity declining to below 0.1 within 12 months. The study introduced the SHAP (SHapley Additive exPlanations) value interpretation framework to clarify the contribution weights of various influencing factors to the model's prediction results, thereby enhancing the transparency of the model's decision-making process. This research will conduct internal validation and iterative optimization of the models by including newly admitted patients, ultimately developing an ARN precision diagnosis and treatment assistance tool suitable for clinical practice. This tool aims to effectively reduce the risk of vision loss caused by ARN through personalized treatment recommendations and prognostic early warning, and promote the standardization of diagnosis and treatment processes. It also provides a reference paradigm for the precise management of similar diseases in primary medical institutions, possessing significant clinical significance and application value.

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

1. Jointly diagnosed by ≥2 fundus disease specialists with associate chief physician title, meeting internationally recognized ARN diagnostic criteria (referring to Takase H criteria 2015 and Uveitis Working Group criteria 2021), specifically satisfying at least 2 of the 3 core indicators: (1) peripheral retinal necrosis lesions, (2) occlusive retinal arteritis, (3) vitreous opacity, accompanied by positive HSV/VZV detection in intraocular fluid PCR or characteristic disease progression manifestations;
2. Visited Shanxi Eye Hospital with baseline data including 23 core indicators (demographic characteristics, etiological typing, retinal vascular occlusion range, treatment plan, etc.), with key data missing rate <=5%;
3. Informed consent obtained from patients or legal guardians, able to cooperate with ≥ 3 months of follow-up to collect prognostic data such as retinal detachment and visual acuity changes;
4. No restrictions on age or gender, including patients with monocular/binocular involvement, covering different disease severities (mild/moderate/severe) and etiological types (VZV/HSV infection) to ensure sample representativeness.

排除标准：

1. 不符合 ARN 诊断标准，或确诊为易混淆疾病（如 CMV 性视网膜炎、弓形虫性视网膜脉络膜炎、增殖性糖尿病视网膜病变等）； 2. 核心指标（病原学结果、治疗方案、视力监测等）缺失率＞5%，且无法通过临床记录补充； 3. 合并严重心肝肾衰竭、晚期恶性肿瘤等，预期生存期＜3 个月，或因精神障碍无法配合随访； 4. 既往接受角膜移植、青光眼滤过术等眼部手术（单纯白内障术后恢复良好者除外），可能影响诊疗及预后评估； 5. 拒绝提供临床数据用于模型构建，或不同意参与随访及验证流程。

Exclusion criteria：

1. Not meeting ARN diagnostic criteria, or diagnosed with confused diseases (e.g., CMV retinitis, toxoplasmic retinochoroiditis, proliferative diabetic retinopathy, etc.); 2. Missing rate of core indicators (etiological results, treatment plan, visual acuity monitoring, etc.) >5%, which cannot be supplemented by clinical records; 3. Complicated with severe heart/liver/kidney failure, advanced malignancy, etc., with expected survival < 3 months, or unable to cooperate with follow-up due to mental disorders; 4. Previous ocular surgeries (except uncomplicated cataract surgery with good recovery) such as corneal transplantation and glaucoma filtration surgery, which may affect diagnosis, treatment and prognostic evaluation; 5. Refusing to provide clinical data for model construction, or disagreeing to participate in follow-up and verification procedures.

研究实施时间：

Study execute time：

从 From 2025-09-01 00:00:00至 To 2027-09-30 00:00:00

征募观察对象时间：

Recruiting time：

从 From 2025-10-30 00:00:00 至 To 2026-01-30 00:00:00

干预措施：

Interventions：

组别：	AI 辅助诊疗组	样本量：	20
Group：	AI-Assisted Diagnosis and Treatment Group	Sample size：	20
干预措施：	诊断阶段：将患者的临床症状、眼底检查结果、病原学检测数据等输入 “急性视网膜坏死综合征 AI 辅助诊疗及预后评估模型”，模型输出 ARN 诊断概率及鉴别诊断建议，辅助临床医师明确诊断。治疗阶段：模型结合患者个体特征（如年龄、合并症、病原学类型等），输出个性化的抗病毒治疗方案（药物选择、剂量、疗程）、玻璃体切除术时机建议及术后管理方案等，临床医师参考模型建议制定最终治疗方案。预后评估阶段：模型根据患者治疗过程中的各项指标变化，动态预测 3 个月内视网膜脱离风险、视力预后等，为患者随访及后续干预提供依据。	干预措施代码：
Intervention：	Diagnostic stage: Input the patient's clinical symptoms, fundus examination results, etiological detection data, etc. into the "Acute Retinal Necrosis Syndrome AI-Assisted Diagnosis, Treatment and Prognostic Evaluation Model". The model outputs the ARN diagnosis probability and differential diagnosis suggestions to assist clinicians in confirming the diagnosis. Treatment stage: Combined with the patient's individual characteristics (such as age, comorbidities, etiological type, etc.), the model outputs personalized antiviral treatment plans (drug selection, dosage, course of treatment), suggestions on the timing of vitrectomy, postoperative management plans, etc. Clinicians refer to the model's suggestions to formulate the final treatment plan. Prognostic evaluation stage: Based on the changes of various indicators during the patient's treatment, the model dynamically predicts the risk of retinal detachment and visual prognosis within 3 months, providing a basis for patient follow-up and subsequent intervention.	Intervention code：

研究实施地点：

Countries of recruitment and research settings：

国家：	中国	省(直辖市)：	山西	市(区县)：	太原
Country：	China	Province：	Shanxi	City：	Taiyuan
单位(医院)：	山西省眼科医院	单位级别：	三甲
Institution hospital：	Shanxi Eye Hospital	Level of the institution：	Tertiary A

测量指标：

Outcomes：

指标中文名：	视网膜脱离发生率	指标类型：	主要指标
Outcome：	Incidence of retinal detachment	Type：	Primary indicator
测量时间点：		测量方法：
Measure time point of outcome：		Measure method：

指标中文名：	玻璃体炎症分级	指标类型：	次要指标
Outcome：	Vitreous inflammation grading	Type：	Secondary indicator
测量时间点：		测量方法：
Measure time point of outcome：		Measure method：

指标中文名：	症状至抗病毒治疗时间	指标类型：	次要指标
Outcome：	Time from symptom onset to antiviral treatment	Type：	Secondary indicator
测量时间点：		测量方法：
Measure time point of outcome：		Measure method：

指标中文名：	合并症情况	指标类型：	次要指标
Outcome：	Comorbidity status	Type：	Secondary indicator
测量时间点：		测量方法：	记录患者是否合并糖尿病、高血压、自身免疫性疾病等其他疾病，分析合并症对 ARN 诊疗及预后的影响，助力 AI 模型更精准地进行个体化诊疗。
Measure time point of outcome：		Measure method：

指标中文名：	最佳矫正视力	指标类型：	主要指标
Outcome：	Best Corrected Visual Acuity	Type：	Primary indicator
测量时间点：		测量方法：
Measure time point of outcome：		Measure method：

指标中文名：	累及第二只眼情况	指标类型：	主要指标
Outcome：	Involvement of the second eye	Type：	Primary indicator
测量时间点：		测量方法：
Measure time point of outcome：		Measure method：

指标中文名：	基线特征指标（如年龄、性别、患眼等）	指标类型：	主要指标
Outcome：	Baseline characteristics (e.g., age, gender, affected eye)	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：	无	组织：
Sample Name：	None	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	12	岁 years
最大 Max age	80	岁 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）：

无

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：	无
Blinding：	None
试验完成后的统计结果（上传文件）：	点击下载
Calculated Results after the Study Completed(upload file)：	download

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）：	无
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：	暂未确定/Not yet
注册人： Name of Registration：	2025-10-27 10:04:10