Automatic Recognition of Laryngoscopic Images Using a Deep‐Learning Technique. (18th February 2020)

• Record Type: Journal Article
• Authors: Ren, Jianjun; Jing, Xueping; Wang, Jing; Ren, Xue; Xu, Yang; Yang, Qiuyun; Ma, Lanzhi; Sun, Yi; Xu, Wei; Yang, Ning; Zou, Jian; Zheng, Yongbo; Chen, Min; Gan, Weigang; Xiang, Ting; An, Junnan; Liu, Ruiqing; Lv, Cao; Lin, Ken; Zheng, Xianfeng; Lou, Fan; Rao, Yufang; Yang, Hui; Liu, Kai; Liu, Geoffrey; Lu, Tao; Zheng, Xiujuan; Zhao, Yu
• Is Part Of: Laryngoscope. Volume 130:Number 11(2020)
• Page Start: E686
• Page End: E693
• Publication Date: 2020-02-18
• Subjects: Deep learning -- laryngoscopic image -- artificial intelligence -- convolutional neural networks -- clinical visual assessment.; Otolaryngology -- Periodicals; 617.51005
• DOI: 10.1002/lary.28539
• Languages: English
• ISSNs: 0023-852X

Abstract

Abstract : Objectives/Hypothesis: To develop a deep‐learning–based computer‐aided diagnosis system for distinguishing laryngeal neoplasms (benign, precancerous lesions, and cancer) and improve the clinician‐based accuracy of diagnostic assessments of laryngoscopy findings. Study Design: Retrospective study. Methods: A total of 24, 667 laryngoscopy images (normal, vocal nodule, polyps, leukoplakia and malignancy) were collected to develop and test a convolutional neural network (CNN)‐based classifier. A comparison between the proposed CNN‐based classifier and the clinical visual assessments (CVAs) by 12 otolaryngologists was conducted. Results: In the independent testing dataset, an overall accuracy of 96.24% was achieved; for leukoplakia, benign, malignancy, normal, and vocal nodule, the sensitivity and specificity were 92.8% vs. 98.9%, 97% vs. 99.7%, 89% vs. 99.3%, 99.0% vs. 99.4%, and 97.2% vs. 99.1%, respectively. Furthermore, when compared with CVAs on the randomly selected test dataset, the CNN‐based classifier outperformed physicians for most laryngeal conditions, with striking improvements in the ability to distinguish nodules (98% vs. 45%, P < .001), polyps (91% vs. 86%, P < .001), leukoplakia (91% vs. 65%, P < .001), and malignancy (90% vs. 54%, P < .001). Conclusions: The CNN‐based classifier can provide a valuable reference for the diagnosis of laryngeal neoplasms during laryngoscopy, especially for distinguishing benign, precancerous, and cancer lesions. Level of Evidence: NA Laryngoscope, 130:E686–E693, 2020