An Efficient Gastrointestinal Hemorrhage Detection and Diagnosis Model for Wireless Capsule Endoscopy
R. Ponnusamy1, S. Sathiamoorthy2
1R. Ponnusamy, Department of Computer and Information Science, Annamalai University, India.
2S.Sathiyamoorthy, Tamil Virtual Academy, India.
Manuscript received on 02 August 2019. | Revised Manuscript received on 07 August 2019. | Manuscript published on 30 September 2019. | PP: 7549-7554 | Volume-8 Issue-3 September 2019 | Retrieval Number: C6171098319/19©BEIESP | DOI: 10.35940/ijrte.C6171.098319
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Wireless Capsule Endoscopy (WCE) captures the section of human gastrointestinal (GI) tract which is impossible by the classical endoscopy investigations. A main limitation exist in the method is the requirement of analyzing massive data quantity for detecting the diseases which consumes more time and increases the burden to the physicians. As a result, there is a requirement to effectively develop an automated model to detect and diagnosis diseases on the WCEimages. The design of the presented model depends upon the examination of the patterns exist in frequency spectra of the WCE frames because of the occurrence of bleeding regions. For the exploration of the discriminating patterns,this study presents a new feature extraction based classification model is developed. An efficient Normalized Gray Level Co-occurrence Matrix (NGLCM) is applied for extracting the features of the GI images. Then, a kernel support vector machine (KSVM) with particle swarm optimization (PSO) is applied for the classification of the processed GI images. The experimentation takes place on the benchmark GI images to verify the superior nature of the presented model. The results confirmed the enhanced classifier outcome of the presented model on all the applied images under several aspects.
Keywords: WCE, Feature Extraction, Diagnosis, Classification
Scope of the Article: Classification