Automatic recognizing of vocal fold disorders from glottis images

Abstract

The laryngeal video stroboscope is an important instrument to test glottal diseases and read vocal fold images and voice quality for physician clinical diagnosis. This study is aimed to develop a medical system with functionality of automatic intelligent recognition of dynamic images. The static images of glottis opening to the largest extent and closing to the smallest extent were screened automatically using color space transformation and image preprocessing. The glottal area was also quantized. As the tongue base movements affected the position of laryngoscope and saliva would result in unclear images, this study used the gray scale adaptive entropy value to set the threshold in order to establish an elimination system. The proposed system can improve the effect of automatically captured images of glottis and achieve an accuracy rate of 96%. In addition, the glottal area and area segmentation threshold were calculated effectively. The glottis area segmentation was corrected, and the glottal area waveform pattern was drawn automatically to assist in vocal fold diagnosis. When developing the intelligent recognition system for vocal fold disorders, this study analyzed the characteristic values of four vocal fold patterns, namely, normal vocal fold, vocal fold paralysis, vocal fold polyp, and vocal fold cyst. It also used the support vector machine classifier to identify vocal fold disorders and achieved an identification accuracy rate of 98.75%. The results can serve as a very valuable reference for diagnosis.

Videomics: bringing deep learning to diagnostic endoscopy

“…This is due to use of a standardized endoscopic approach through trans-nasal or transoral video-endoscopy and the relative similarity with gastrointestinal subsites. In 2014, Huang et al [31] proposed an automatic system aimed at recognizing images of the glottis and classifying different vocal fold disorders. The technique was based on a support vector machine classifier and reached an accuracy of 99%.…”

Artificial intelligence in clinical endoscopy: Insights in the field of videomics

“…Other studies ( 12 , 13 ) have employed ML to classify pharyngo-laryngeal benign lesions during videoendoscopy, demonstrating notable results. A preliminary attempt was described in 2014 by Huang et al ( 12 ), who proposed an automatic system aimed at recognizing the dynamic image of the glottis and classifying different vocal fold disorders (“normal vocal fold,” “vocal fold paralysis,” “vocal fold polyp,” and “vocal fold cyst”). This study used an SVM classifier and reached an accuracy of 98.7%.…”

A Deep Learning Enhanced Novel Software Tool for Laryngeal Dynamics Analysis

“…Many works focused on providing an automatic approach to segment the glottal area (Gloger et al, 2015;Huang et al, 2014;Matsuo et al, 2018;Poznyakovskiy et al, 2015). For instance, these are based on edge detection (Pinheiro et al, 2014), Bayes-supported level set segmentation (Gloger et al, 2015), Gabor filtering together with principal component analysis (Mendez-Zorrilla & Garcia-Zapirain, 2015), and active contours (Ammar, 2019).…”

“…However, the segmentation of the glottal area has been shown to be a difficult task requiring varying degrees of user interaction (Andrade- Miranda et al, 2020;Andrade-Miranda & Godino-Llorente, 2017;Gloger et al, 2015;Huang et al, 2014;Lohscheller et al, 2007;Mendez-Zorrilla & Garcia-Zapirain, 2015;Naghibolhosseini et al, 2018;Osma-Ruiz et al, 2008;Shi et al, 2015). These algorithms were also only validated on very limited data sets.…”