|
|
Bimonthly Since 1986 |
ISSN 1004-9037
|
|
|
|
|
Publication Details |
Edited by: Editorial Board of Journal of Data Acquisition and Processing
P.O. Box 2704, Beijing 100190, P.R. China
Sponsored by: Institute of Computing Technology, CAS & China Computer Federation
Undertaken by: Institute of Computing Technology, CAS
Published by: SCIENCE PRESS, BEIJING, CHINA
Distributed by:
China: All Local Post Offices
|
|
|
|
|
|
|
|
|
|
Abstract
The complexity of the brain's image and structure makes the analysis of brain tumors difficult, making brain tumor extraction and analysis difficult jobs in medical image processing. An important part of medical image processing is performed through segmentation and classification. When it comes to diagnosing conditions of the brain and other organs, MRI (magnetic resonance imaging) has become an invaluable tool. This research effort analyzes the similarities and differences between two different classifiers used for tumor diagnosis. Some of the techniques used include “gray level co-occurrence matrix (GLCM)” feature extraction, “principal component analysis” and ostu and morphological segmentation. In order to identify tumors, features are classified with a “support vector machine (SVM)” and a “convolutional neural network (CNN)”. Ultimately, the tumor's location and form are pinpointed by parsing the MR picture. The experimental findings show that CNN classification was more successful than SVM classification (99.67%). When a brain tumor is diagnosed, its seriousness may be determined at last.
Keyword
Brain tumor, Support Vector Machine, Convolution Neural Network, Magnetic resonance imaging, Ostu.
PDF Download (click here)
|
|
|
|
|