PCA-SVM based CAD System for Focal Liver Lesions using B-Mode Ultrasound Images

  • Jitendra Virmani Indian Institute of Technology Roorkee, Roorkee
  • Vinod Kumar Indian Institute of Technology Roorkee, Roorkee
  • Naveen Kalra Post Graduate Institute of Medical Education and Research, Chandigarh
  • Niranjan Khandelwa Post Graduate Institute of Medical Education and Research, Chandigarh
Keywords: Focal liver lesions, B-mode ultrasound, principal component analysis, support vector machine classifier, computer-aided diagnostic system

Abstract

The contribution made by texture of regions inside and outside of the lesions in classification of focal liver lesions (FLLs) is investigated in the present work. In order to design an efficient computer-aided diagnostic (CAD) system for FLLs, a representative database consisting of images with (1) typical and atypical cases of cyst, hemangioma (HEM) and metastatic carcinoma (MET) lesions, (2) small as well as large hepatocellular carcinoma (HCC) lesions and (3) normal (NOR) liver tissue is used. Texture features are computed from regions inside and outside of the lesions. Feature set consisting of 208 texture features, (i.e. 104 texture features and 104 texture ratio features) is subjected to principal component analysis (PCA) for finding the optimal number of principal components to train a support vector machine (SVM) classifier for the classification task. The proposed PCA-SVM based CAD system yielded classification accuracy of 87.2% with the individual class accuracy of 85%, 96%, 90%, 87.5% and 82.2% for NOR, Cyst, HEM, HCC and MET cases respectively. The accuracy for typical, atypical, small HCC and large HCC cases is 87.5%, 86.8%, 88.8%, and 87% respectively. The promising results indicate usefulness of the CAD system for assisting radiologists in diagnosis of FLLs.

Defence Science Journal, 2013, 63(5), pp.478-486, DOI:http://dx.doi.org/10.14429/dsj.63.3951

Author Biographies

Jitendra Virmani, Indian Institute of Technology Roorkee, Roorkee

Mr Jitendra Virmani did his BTech (Hons.) in Instrumentation Engineering from Sant Longowal Institute of Engineering and Technology in 1999 and MTech in Measurement and Instrumentation from IIT-Roorkee in 2006. Currently pursuing his PhD at Biomedical Instrumentation Laboratory, Electrical Engineering Department, IIT-Roorkee as a full time Research Scholar under MHRD Assistantship. He is Life member of Institute of Engineers (IEI), India. His research interests include: Application of machine learning and soft computing techniques for analysis of medical images.

Vinod Kumar, Indian Institute of Technology Roorkee, Roorkee
Dr Vinod Kumar obtained BSc (Electrical Engineering) with from Punjab University in 1973, ME (Measurement & Instrumentation) and PhD from University of Roorkee in 1975 and 1984, respectively. Currently working as a Professor, Department of Electrical Engineering, IIT Roorkee since 1995. He has guided 21 Doctoral and more than 80 Master’s Thesis and has more than 150 research publications in internationally reputed journals and conference proceedings. He is a senior member of IEEE, USA. His research interests include : Medical image processing, digital signal processing and telemedicine.
Naveen Kalra, Post Graduate Institute of Medical Education and Research, Chandigarh
Dr Naveen Kalra did MBBS and MD from Maulana Azad Medical College, New Delhi in 1992 and 1998 respectively. Currently working as a Additional Professor at Department of Radio diagnosis and Imaging, Post Graduate Institute of Medical Education and Research, Chandigarh. He has guided 30 MD Thesis and published more than 85 research publications and contributed 15 chapters in books.
Niranjan Khandelwa, Post Graduate Institute of Medical Education and Research, Chandigarh
Dr Niranjan Khandelwal did his MBBS from Calcutta University in 1980 and MD (Radiodiagnosis) from Postgraduate Institute of Medical Education & Research, Chandigarh in 1984. He is Professor & Head, Department of Radiodiagnosis and Imaging, PGIMER- Chandigarh. He has published more than 140 research papers and contributed 15 chapters in books.
Published
2013-09-25
How to Cite
VirmaniJ., KumarV., KalraN., & KhandelwaN. (2013). PCA-SVM based CAD System for Focal Liver Lesions using B-Mode Ultrasound Images. Defence Science Journal, 63(5), 478-486. https://doi.org/10.14429/dsj.63.3951
Section
Biomedical Sciences