Using Cluster Analysis for Image Processing in High Speed Video Laryngoscopy

Abstract

This paper summarizes findings related to the problematics of glottis detection in video sequences obtained by medical examination of vocal cords by high speed videolaryngoscopy (HSV). The glottis detection is based on cluster analysis method K-means which complements the existing set of segmentation methods used in detection algorithms. This method has been tested on a large corpus of HSV sequences from clinical practice on ENT department.