Detection of Liver Lesion using Robust Machine Learning Technique
Sowmya Sundari L K1, Nirmala S Guptha2, Shruthi G3, Thanuja K4, Anitha K5
1Sowmya Sundari L K, Department of C & IT, REVA University, Bangalore (Karnataka), India.
2Dr. Nirmala S Guptha, Department of C & IT, REVA University, Bangalore (Karnataka), India.
3Shruthi G, Department of C & IT, REVA University, Bangalore (Karnataka), India.
4Thanuja K, Department of C & IT, REVA University, Bangalore (Karnataka), India.
5Anitha K, Department of C & IT, REVA University, Bangalore (Karnataka), India.
Manuscript received on 05 June 2019 | Revised Manuscript received on 14 June 2019 | Manuscript Published on 29 June 2019 | PP: 214-219 | Volume-8 Issue-5S, May 2019 | Retrieval Number: E10440585S19/19©BEIESP
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 (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: In the present era, Computer Aided Diagnosis (CAD) is very useful for the detection of a liver tumor. This type of study and categorization system can moderate an unnecessary biopsy. The proposed method for the detection of liver cancer clusters in liver images using Gabor features and shape features. The mentioned regions are categorized by the SVM classifier utilizing the most prevailing features selected from the above features. In our project, we have proposed a systematic approach of analyzing a liver under cancer positive environment. We have proposed a technique for tumor identification and segmentation using image smoothing and refining methods. When we use CT images for the detection of liver tumor manual interaction is not necessary, since it works automatically. The projected method needs to learn a few model parameters such as tumor part, non-tumor part, and segment liver regions. The complete system is divided into the training part and testing part respectively and this system is based mainly on SVM. The input liver image undergoes for the preprocessing step and image segmentation. Preprocessing includes many steps like the resizing of an image, improve the clarity of the image, conversion form colored image into grayscale. After these necessary features are collected from the resulting image. These collected features are then fed to the SVM for training. These collected features are compared with examination results by the SVM Classifier with the existing trained features using RBF kernel. Contingent upon the correlation result, the classifier gives the outcome.
Keywords: Machine Learning Segmentation Image Classifier.
Scope of the Article: Machine Learning