期刊文献+

基于荧光素眼底血管造影的急性期早产儿视网膜病变治疗分类模型的建立与验证 被引量:2

Development and validation of treatment classification model for acute-phase retinopathy of prematurity based on fundus fluorescein angiography
下载PDF
导出
摘要 目的 探讨急性期早产儿视网膜病变(ROP)患者荧光素眼底血管造影(FFA)检查中与需要治疗相关的指标,建立相应治疗分类模型并加以验证。方法 回顾性纳入2018年4月至2022年1月在空军军医大学西京医院眼科行全身麻醉下FFA检查的急性期ROP患儿114例227眼。患儿按73的比例随机分为训练集(80例160眼)和验证集(34例67眼)。训练集中,38例76眼患儿接受激光治疗(治疗组),42例84眼患儿未治疗发生自然退行(非治疗组);验证集中,15例29眼患儿接受治疗,19例38眼患儿随诊观察。纳入后极部视网膜血管形态和血管-无血管交界处、血管化区域内改变及黄斑区的FFA特征,借助视盘中央与最靠近后极部的病变边缘之间的距离(DB)和视盘中央与黄斑中心凹距离(DF)的比值(DB/DF)评估病变位置。在训练集中通过单因素和多因素Logistic回归分析ROP需要治疗的独立危险因素,建立治疗分类模型并绘制列线图。在训练集和验证集中应用受试者工作特征曲线下面积(AUC)评估模型的区分度。结果 训练集中,治疗组与非治疗组患儿视网膜血管形态、荧光素渗漏分布比较,差异均有统计学意义(均为P<0.05)。治疗组患儿嵴后血管迂曲扩张、“爆米花”病变、异常血管分支和毛细血管床的发生率均高于非治疗组,颞侧、鼻侧DB/DF值均小于非治疗组,差异均有统计学意义(均为P<0.05)。训练集和验证集患儿分别有10眼(6.25%)和11眼(16.42%)在血管化区域出现弱荧光区域,差异有统计学意义(P=0.016)。单因素Logistic回归分析结果显示,后极部视网膜血管形态、嵴后血管迂曲扩张、荧光素渗漏、“爆米花”病变、异常血管分支、毛细血管床丢失和颞侧、鼻侧DB/DF值与ROP需要治疗相关(P<0.1)。多因素Logistic回归分析结果显示,后极部视网膜血管形态(轻度迂曲扩张:P=0.001;血管迂曲扩张:P<0.001)、嵴后血管迂曲扩张(P=0.002)、荧光素渗� Objective To explore the indicators related to the need for treatment of acute-phase retinopathy of prematurity(ROP)in fundus fluorescein angiography(FFA),and to develop and validate the treatment classification model.Methods A total of 227 eyes of 114 infants with acute-phase ROP who underwent FFA under general anesthesia from April 2018 to January 2022 in the Department of Ophthalmology of Xijing Hospital Affiliated to Air Force Medical University were retrospectively included.The infants were randomly divided into the training set(160 eyes of 80 infants),and the validation set(67 eyes of 34 infants)at a ratio of 7 GA6FA 3.In the training set,76 eyes of 38 patients underwent the laser therapy(treatment group),while 84 eyes of 42 patients regressed spontaneously in the context of no treatment(non-treatment group).In the validation set,29 eyes of 15 patients received therapy,and 38 eyes of 19 patients were under follow-up observation.Retinal vessel morphology of the posterior pole and FFA characteristics at the vascular-avascular junction,within the vascularized zone,and in the macular area were collected.The location of the lesion was determined by the ratio between the distance from the center of the optic disc to the border of the lesion closest to the posterior pole(DB)and the distance from the center of the optic disc to the macular fovea(DF)(DB/DF).In the training set,univariate and multivariate logistic regression analyses were performed to determine the independent risk factors for treatment-requiring ROP,and the treatment classification model and the nomogram were further developed.The receiver operating characteristic curve and area under the curve(AUC)was used to evaluate the model discrimination in the training set and validation set.Results The differences in the retinal vessel morphology and distribution of fluorescein leakage were statistically significant between the treatment group and the non-treatment group in the training set(both P<0.05).The incidences of vessel dilation and tortuosity of the
作者 武雷 李曼红 钱迪 严宏祥 王亮 周子义 周毅 樊静 苟凯丽 王雨生 张自峰 WU Lei;LI Manhong;QIAN Di;YAN Hongxiang;WANG Liang;ZHOU Ziyi;ZHOU Yi;FAN Jing;GOU Kaili;WANG Yusheng;ZHANG Zifeng(Department of Ophthalmology,Eye Institute of Chinese PLA,Xijing Hospital,Fourth Military Medical University,Xi’an 710032,Shaanxi Province,China;Department of Health Statistics,Faculty of Health Services of Naval Medical University,Shanghai 200433,China)
出处 《眼科新进展》 CAS 北大核心 2023年第4期284-289,共6页 Recent Advances in Ophthalmology
基金 国家自然科学基金(编号:81770936) 陕西省重点研发计划(编号:2021SF-159) 西京医院临床应用研究课题(编号:JSYXM02)。
关键词 荧光素眼底血管造影 早产儿视网膜病变 治疗分类模型 列线图 fundus fluorescein angiography retinopathy of prematurity severity of illness treatment nomogram
  • 相关文献

参考文献8

二级参考文献62

共引文献300

同被引文献18

引证文献2

  • 1夏平,何志豪,雷帮军,彭程,王雨蝶.全尺度密集卷积U型网络的视网膜血管分割算法[J].计算机工程与设计,2024,45(3):866-873.
  • 2《人工智能在视网膜图像自动分割和疾病诊断中的应用指南(2024)》专家组,国际转化医学会眼科专业委员会,中国医药教育协会眼科影像与智能医疗分会,中国眼科影像研究专家组,邵毅,张铭志,许言午,迟玮,刘祖国,谭钢,陈有信,杨卫华,接英,张慧,李世迎,廖萱,邵婷婷,计丹,马健,杨文利,田磊,胡亮,蔡建奇,彭娟,陆成伟,肖鹏,刘光辉,苏兆安,董诺,秦牧,李程,邹文进,刘籦,赵慧,陈新建,陈琦,文丹,黄明海,温鑫,李中文,石文卿,顾正宇,董贺,唐丽颖,蒋贻平,宋秀胜,王遷,葛倩敏,邱坤良,李正日,刘秋平,易湘龙,康刚劲.人工智能在视网膜图像自动分割和疾病诊断中的应用指南(2024)[J].眼科新进展,2024,44(8):592-601.

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部