A methodological approach to classify typical and atypical pigmentnetwork patterns for melanoma diagnosis

Pathan, Sameena and Prabhu, G K and Siddalingaswamy, PC A methodological approach to classify typical and atypical pigmentnetwork patterns for melanoma diagnosis. Biomedical Signal Processing and Control, 44 (03). pp. 25-37. ISSN 1746-8094

[img] PDF
4573.pdf - Published Version
Restricted to Registered users only

Download (4MB) | Request a copy

Abstract

The pigment network is considered as one of the most histopathologically relevant indicator of melanoma.The objective of this empirical study is to design a novel automatic system for detection of pigmentnetwork and provide a differentiation between typical and atypical network patterns. The algorithmdesign consists a set of sequential stages. Pigment network masks are detected using a bank of 2D Gaborfilters, and a set of pigment network features are extracted to determine the role of pigment network inthe diagnosis of the lesion. In the second stage, a machine learning process is carried out using the rulesgenerated from the pigment network masks to identify the typical and atypical pigment network patterns.The proposed methodology was tested on the PH2dataset of 200 images, obtaining an average sensitivityof 96%, specificity of 100% and accuracy of 96.7% for lesion diagnosis, and an average sensitivity, specificityand accuracy of 84.6%, 88.7% and 86.7% respectively, for pigment network classification. The proposedsystem stands out amongst the few state of art literatures reported in the context of dermoscopic imageanalysis in terms of performance and methodologies adopted, thus proving the reliability of the proposedstudy.

Item Type: Article
Uncontrolled Keywords: Atypical, Dermoscopy, Gabor, Polynomial curve fitting, Pigment network, Typical
Subjects: Engineering > MIT Manipal > Computer Science and Engineering
Engineering > MIT Manipal > Electronics and Communication
Depositing User: MIT Library
Date Deposited: 03 Jul 2018 04:44
Last Modified: 03 Jul 2018 04:44
URI: http://eprints.manipal.edu/id/eprint/151454

Actions (login required)

View Item View Item