MRI Brain Tumor Texture based Segmentation and Classification Using Ensemble Classifier

Author (s) : B. Vijaya Kumar and Dr. Parasuraman Kumar

---

Abstract :
Brain tumor is a band of tissue. It is pre organized through a less addition of not regular cells. It happens the cell obtains abnormal from what's usually predictable creation and construction group of objects inside the brain not very long ago it is becoming an important issue of death of more number of people. The importance of brain tumor has massive huge among overall the differentiations of cancers, hence to save a life sudden detection and correct treatment has to be made. It is vital extremely crucial to make comparisons brain tumor from the MRI treatment. It is extremely hard to have clear about the different from what are usually likely structures of human brain using easily imaging ways of doing things. The discussed system consists of four process are done to identify the brain tumors. The primary system is pre-processing the image data from the gathering of computer file full of information using median filtering, second stage is division of something into smaller parts using Fuzzy C-means Clustering and LIPC, third stage is feature extraction through the Gobar filter and the fourth stage is classification through the group of performers or objects variations is the combination of FFANN, Extreme Learning Machine (ELM) and Support Vector Machine classifier (SVM). Experiments have exposed that the system was more strong and healthy to initialization, quicker and very close to the truth or true number.

Keywords : Ensemble Classifiers, Gobar filter, LIPC, ELM, SVM, Feed Forward ANN and Fuzzy C-means Clustering.

Conference Date : 26-04-2019

Pages : 5-13

Paper ID : 3218153