Open Access iconOpen Access

ARTICLE

crossmark

Classification of Liver Tumors from Computed Tomography Using NRSVM

S. Priyadarsini1,*, Carlos Andrés Tavera Romero2, M. Mrunalini3, Ganga Rama Koteswara Rao4, Sudhakar Sengan5

1 Department of Computer Science and Engineering, P. S. R. Engineering College, Sivakasi, Tamil Nadu, 626123, India
2 COMBA R&D Laboratory, Faculty of Engineering, Universidad Santiago de Cali, 76001, Cali, Colombia
3 Master of Computer Applications, M S Ramaiah Institute of Technology, Bangalore, 560054, India
4 Department of Computer Science & Information Technology, Koneru Lakshmaiah Education Foundation, Vaddeswaram, A.P, India
5 Department of Computer Science and Engineering, PSN College of Engineering and Technology, Tirunelveli, 627152, Tamil Nadu, India

* Corresponding Author: S. Priyadarsini. Email: email

Intelligent Automation & Soft Computing 2022, 33(3), 1517-1530. https://doi.org/10.32604/iasc.2022.024786

Abstract

A classification system is used for Benign Tumors (BT) and Malignant Tumors (MT) in the abdominal liver. Computed Tomography (CT) images based on enhanced RGS is proposed. Diagnosis of liver diseases based on observation using liver CT images is essential for surgery and treatment planning. Identifying the progression of cancerous regions and Classification into Benign Tumors and Malignant Tumors are essential for treating liver diseases. The manual process is time-consuming and leads to intra and inter-observer variability. Hence, an automatic method based on enhanced region growing is proposed for the Classification of Liver Tumors (LT). To enhance the Liver Region (LR) from the surrounding tissues, Non-Linear Mapping (NLP) is used. Region Growing Segmentation (RGS) is employed to segment the LR, and Expectation-Maximization (EM) algorithm is used to segment the region of interest. Grey Level Co-occurrence Matrix (GLCM) features are extracted from the tumor region, and Nonlinear Random Support Vector Machine (NRSVM) classification is performed to classify the Benign Tumors and Malignant Tumors. The proposed method is tested on a database of medical images collected from Med all Diagnostic Research Centre and attained an accuracy of 96%. The proposed method is beneficial for better liver tumor diagnosis in an optimized method by the medical expert.

Keywords


Cite This Article

S. Priyadarsini, C. Andrés Tavera Romero, M. Mrunalini, G. Rama Koteswara Rao and S. Sengan, "Classification of liver tumors from computed tomography using nrsvm," Intelligent Automation & Soft Computing, vol. 33, no.3, pp. 1517–1530, 2022. https://doi.org/10.32604/iasc.2022.024786



cc This work is licensed under a Creative Commons Attribution 4.0 International License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
  • 1205

    View

  • 628

    Download

  • 0

    Like

Share Link