Performance Analysis of Classification of Liver Tumors using Support Vector Machine and Rough Set based Classifiers
Aravinda H L1, Sudhamani M V2

1Mr. Aravinda H L, Research Scholar, Assistant Professor, Department of Telecommunication Engineering, Dr. Ambedkar Institute of Technology, Bengaluru (Karnataka), India.

2Dr. M V Sudhamani, Professor & HOD, Department of Information Secience & Engineering, RNS Institute of Technology, Bengaluru (Karnataka), India.

Manuscript received on 03 December 2019 | Revised Manuscript received on 11 December 2019 | Manuscript Published on 31 December 2019 | PP: 99-102 | Volume-9 Issue-2S December 2019 | Retrieval Number: B10621292S19/2019©BEIESP | DOI: 10.35940/ijitee.B1062.1292S19

Open Access | Editorial and Publishing 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 (

Abstract: In recent years the medical diagnosis is majorly done based on the medical images captured using various imaging modalities. The medical doctors and radiologists use these medical images to identify the pathological problems or diseases and suggest the patient about further treatment. In this process, medical doctors and radiologists often prefer to make use of software which can assist in taking the decision. Such an approach is called as Computer Aided Diagnosis (CAD). The CAD systems usually comprise of many phases like segmentation of portion corresponding to a particular organ or region under consideration, detecting the pathology bearing area in that and further classifying into various disease classes. Here, Accuracy of classifiers majorly decides the effectiveness of the diagnosis. In this paper, classification of liver tumors into benign and malignant is considered as a case study. Implementation of two different classifiers namely Support Vector Machine and Rough Set based classifier is carried out using set of features extracted such as Texture features using Average Correction Higher order Local Autocorrelation Coefficients and shape features using Legendre moments. Comparison of performance of both the classifiers is made and tabulated. Here, Rough Set based classifier has performed better when compared with Support Vector Machine.

Keywords: Liver Tumor, Average Correction Higher order Local Autocorrelation Coefficients, Legendre Moments, Support Vector Machine Classifier, Rough Set based Classifier.
Scope of the Article: Machine Learning