深度学习辅助电子喉镜诊断喉白斑的应用研究  

Application of deep learning assisted electronic laryngoscope in diagnosis of laryngeal leukoplakia

作　　者：付嘉 李丽娟[1] 闫燕[1] 马芙蓉[1] FU Jia;LI Lijuan;YAN Yan

机构地区：[1]北京大学第三医院耳鼻咽喉科,北京100191

出　　处：《临床耳鼻咽喉头颈外科杂志》2021年第5期464-467,共4页Journal of Clinical Otorhinolaryngology Head And Neck Surgery

摘　　要：喉白斑主要表现为喉部黏膜处难以擦去的局限白色病灶,因其多发生于声带黏膜处,又称为声带白斑。自Schwimmer(1877)首先将"白斑"一词用于描述口腔黏膜不同位置的白色病变以来,人们对喉白斑的认识不断加深,Durant(1880)、Pierce(1920)、Jackson(1923)先后对其进行描述和定义。喉白斑的发生发展与多种致病因素的长期作用有关,有一定的恶变倾向,但既往将喉白斑完全等同于癌前病变这一观念并不准确。In recent years,medical imaging technology and computer technology have made great progress.On the one hand,with the development and popularization of electronic laryngoscope,the image of electronic laryngoscope plays a very important role in the diagnosis of vocal cord lesions.On the other hand,deep learning algorithm,especially convolutional neural networkhas gradually become the first choice of medical image recognition since the foundation of deep learning algorithm.So far,deep learning algorithm has made great contributions in many disciplines.In this paper,the basic concept of deep learning,the current status of image recognition of vocal cord lesions,and the prospect of research based on deep learning in vocal cord image lesions recognition are reviewed.

关 键 词：喉白斑 深度学习 电子喉镜 图像识别 

分 类 号：R767.1[医药卫生—耳鼻咽喉科]