Paper Title
Identification and Classification of Vocal Cord Tumours

Many digital image processing techniques are used in the medical practices for image analysis. The commonly found abnormalities in endoscopic images are cancer tumors, ulcers, bleeding due to internal injuries, etc. For laryngeal tumor detection, detecting lesions in the larynx at early stages is one of the most important factors involved in successful disease treatment. In order to obtain the hispathology of an abnormality, the analysis of endoscopic images is one of the most accessible methods. However standard imaging techniques such as white light endoscopy offer limited information about the laryngeal tissue. NBI(Narrowband Imaging) is an alternative to achieve this goal. NBI is an optical technology that enhances the practitioners capability to detect and diagnose lesion through endoscopic inspection. For detecting laryngeal tumors without biopsy and pathological examination, an automatic method which is based on anisotropic filtering and matched filter is described. Lesion classification is then performed using SVM. Here a novel processing framework is presented for the automatic detection and classification of lesions based on segmentation and analysis of blood vessel networks on endoscopic images. Keywords - Blood Vessel Segmentation, Computer-Aided Diagnosis, Laryngoscopy, Segmentation, Lesion Classification, SVM