通讯作者: 常玮, cw262@163.com

DOI：10.12201/bmr.202408.00023

Wang yeyu1,Chang wei2*, Li jun2Corresponding Author: Chang wei(Email: cw262@163.com)

• 摘要：目的:确定内镜白光模式(White Lighting Imaging,WLI)及窄带成像模式(Narrow Band Imaging,NBI)下诊断声带白斑病理类型的最佳截断点,提高无创预测声带白斑病理性质的准确性,为声带白斑患者是否需要进行手术干预及手术的时机提供依据。方法:选择2020年10月至2022年12月就诊于长治医学院附属和平医院耳鼻咽喉头颈外科并行手术明确病理性质的声带白斑(Vocal cord leukoplakia,VCL)患者96例(共122侧病变)。于术前分别进行白光模式及NBI模式下的图像采集,并对病变进行2种分型(白光模式Chen分型、NBI模式Ni分型)的评定。通过比较受试者工作特性曲线(receiver operator characteristic,ROC)的约登指数得出分型(超过2个分层)的最佳截断值。通过比较2种临床分型的曲线下面积(area under curve,AUC)、最佳截断值下的准确度、敏感性、特异性、阳性预测值及阴性预测值评价不同临床分型对声带白斑病理性质的诊断效能。结果:1、白光模式Chen分型、NBI模式Ni分型与病理等级之间均存在相关性(rs=0.513、0.653,P均＜0.001)。2、白光模式Chen分型的最佳截断值为3,即光滑平坦型及光滑隆起型为良性病变,粗糙型为恶性病变；NBI模式Ni分型的最佳截断值为5,即I~IV型为良性病变,V型、VI型为恶性病变。3、白光模式Chen分型、NBI模式Ni分型的AUC分别为0.815,0.875。2种分型(最佳截断值)的准确度、敏感性、特异性、阳性预测值及阴性预测值分别为79.51%,84.62%,75.71%,72.13%,86.89%；85.25%,76.92%,91.42%,86.96%,84.21%。结论:NBI模式Ni分型及白光模式Chen分型的诊断效能均较高,但NBI模式Ni分型(截断值为5)对声带白斑病理性质的诊断效能高于白光模式Chen分型(截断值为3)。

  关键词： 声带白斑,窄带成像模式,白光模式,诊断; ; ; 

   

  Abstract: Objective: The purpose is to select the most accurate classification in White Light Imaging (WLI) and Narrow Band Imaging (NBI) for diagnosing the pathological result of vocal cord leukoplakia and the proposed cut-off point,which can improve the accuracy of non-invasive prediction of the pathological result of vocal cord leukoplakia and can provide basis for whether the vocal cord leukoplakia patients need surgery and the timing of surgical intervention.Methods:96 vocal cord leukoplakia(VCL) patients(122 lesions) were included in the study from October 2021 to December 2022. Before surgery, images were collected in WLI and NBI respectively.According to two classifications (Chen classification in WLI and Ni classification in NBI) ,lesions were evaluated. By comparing Youden’s index,the proposed cut-off point for classification more than two degrees can be assessed.By comparing the area under curve (AUC), accuracy, sensitivity, specificity, positive predictive value and negative predictive value of four clinical classifications per the purposed cut-off point, the diagnostic performance of different clinical classifications on the pathological results of vocal cord leukoplakia was assessed, the most accurate classification and the purposed cut-off point were obtained. Results: 1、Chen classification in WLI was correlated with pathological tissue results(rs=0.609, P＜0.001).Ni classification in NBI was correlated with pathological tissue results(rs=0.763, P＜0.001).2、In WLI,the purposed cut-off point of Chen classification is 3(flat and smooth type is bengin,elecated and smooth type and rough type are maligant)；in NBI,the purposed cut-off point of Ni classification is 5(I~IV tpye are bengin, V and VI type are maligant).3、The AUC of two clinical classifications’ model(Chen classification in WLI and Ni classification in NBI) were 0.815,0.875.The accuracy,sensitivity,specificity,PPV and NPV of the two clinical classifications (purposed cut-off point) were79.51%,84.62%,75.71%,72.13%,86.89%；85.25%,76.92%,91.42%,86.96%,84.21%.Conclusions: Ni classification in NBI (cut-off point 5) and Chen classification(cut-off point 3) both perform well. The diagnostic performance of Ni classification in NBI (cut-off point 5) in WLI is better than Chen classification(cut-off point 3).

  Key words: Vocal; cord leukoplakia, Narrow; band imaging,White; lighting imaging,diagnosis

  提交时间：2024-08-12

  版权声明：作者本人独立拥有该论文的版权，预印本系统仅拥有论文的永久保存权利。任何人未经允许不得重复使用。
• html

• 图表

• 周建国. 大数据背景下结构式医学影像诊断报告的临床应用. 2024. doi: 10.12201/bmr.202409.00066

  布佐拉古丽·图尔荪. 阴道超声与宫腔镜对围绝经期异常子宫出血诊断价值. 2023. doi: 10.12201/bmr.202308.00002

  冯利, 岳小飞. 9种结合式机器学习算法在基于高维数据的肿瘤早期诊断准确性的比较研究. 2021. doi: 10.12201/bmr.202108.00016

  张怡然, 张敬晨, 孙婧雯, 王嘉阳, 张更路, 周伟. 医药卫生与人口健康领域科学数据引用行为与可回溯性研究*. 2024. doi: 10.12201/bmr.202407.00043

  梁静, 文奕. 知识图谱在医学辅助诊断中的应用研究. 2022. doi: 10.12201/bmr.202109.00021

  汤阳杰, 吴晓锋. 腰椎间盘早期退变的智能诊断方法研究. 2022. doi: 10.12201/bmr.202204.00001

  汤阳杰, 吴晓锋, 刘群, 欧阳林, 罗爱芳. 腰椎间盘早期退变的智能诊断方法研究*. 2022. doi: 10.12201/bmr.202112.00007

  刘松, 李渊, 马建岭, 史利卿, 郝素英, 王丽云, 罗景舒, 郭兆迪. 清醒镇静技术在支气管镜检查中的应用研究进展. 2024. doi: 10.12201/bmr.202409.00030

  唐智贤, 李萍, 曾雅楠, 张一乐, 郑宇翔, 李恒峰, 魏千翔, 肖正光. 计算机辅助舌象分析诊断的研究进展. 2021. doi: 10.12201/bmr.202106.00006

  曹丽敏, 张丽敏, 于欣, 王白石, 刘景利. 血清学指标与阴道彩超鉴别诊断早期异位妊娠、先兆流产的价值. 2024. doi: 10.12201/bmr.202408.00036

• 序号	提交日期	编号	操作
  1	2024-07-06	bmr.202408.00023V1	下载

• 公开评论  匿名评论  仅发给作者 提交评论