Medical Image Analysis Using Unsupervised and Supervised Classification Techniques
V. Joseph Peter1, M. Karnan2
1Prof. V. Joseph Peter, Department of Computer Science, Kamaraj College, Tuticorin (Tamil Nadu), India.
2Dr. M.Karnan, Professor and Head, Department of CSE, Tamil Nadu College of Engineering, Coimbatore (Tamil Nadu), India.
Manuscript received on 12 October 2013 | Revised Manuscript received on 20 October 2013 | Manuscript Published on 30 October 2013 | PP: 40-45 | Volume-3 Issue-5, October 2013 | Retrieval Number: E1241103513/13©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: The evolution of digital computers as well as the development of modern theories for learning and information processing leads to the emergence of Computational Intelligence (CI) engineering. Liver surgery remains a difficult challenge in which preoperative data analysis and strategy definition may play a significant role in the success of the procedure. Extraction of liver fibrosis is done using image enhancement techniques using various filtering techniques, unsupervised clustering techniques such as modified k means and fuzzy c means and supervised techniques such as ANN, BPN and feed forward NN. It constructs a statistical model of liver fibrosis from these MRI scans using ANN, SVM, GA with k means, GA with Fuzzy and Feed forward back propagation neural network classifier. Our experimental study analyzed 250 MRI images. These results are better than the existing image-based methods which can only discriminate between healthy and high grade fibrosis subjects. With appropriate extensions, our method may be used for non-invasive classification and progression monitoring of liver fibrosis in human patients instead of more invasive approaches, such as biopsy or contrast-enhanced imaging. The proposed system is tested on more than 300 digitized MRI Image database to establish its competence.
Keywords: Computational Intelligence, Enhancement Techniques, Clustering Techniques, Fuzzy C Means, Back Propagation Neural Network.
Scope of the Article: Classification