|
|
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
Brain tumor (BT) are collections of abnormal brain cells, and they have the potential to cause death. A total of 4.4 per 100,000 men and women died from BT each year. From 2017 to 2019, around 0.6 percent of the population was detected with BT. Thus, identifying BT is crucial for providing effective care to patients. Nowadays, MRI scans are frequently used to diagnose BT. The manual examination may introduce inaccuracy due to the BTs' complexities and peculiarities. The practice of BT identification involves the expertise and experience of doctors; yet not everybody has immediate access to such professionals, therefore automating the practice would be useful. Deep learning (DL) has achieved great levels of effectiveness in identifying tumours from MRI to automate the process. Gray Level Co-occurrence Matrix (GLCM) and Convolution Neural Networks (CNN) are used to identify BT in the proposed work. Features are extracted using GLCM, while classification is performed using CNN. Data from Kaggle is used for both the model's training and its subsequent testing. The proposed approach is compared against a conventional CNN model to determine its effectiveness. The experimental findings indicate that GLCM-CNN is superior to CNN in terms of accuracy
Keyword
Brain, Tumor, Image processing, Resize, Filter, Deep Learning, Accuracy, Epochs, Feature Extraction
PDF Download (click here)
|
|
|
|
|