Wave Deep Neural Network for Multi-Class MRI Brain Image Classification
- Resource Type
- Conference
- Authors
- Ramasamy, Viswanathan; Vinodhini, V.; Saravanan, T. R.; Kumar, S. Dinesh; Lavanya, S.
- Source
- 2023 2nd International Conference on Edge Computing and Applications (ICECAA) Edge Computing and Applications (ICECAA), 2023 2nd International Conference on. :686-690 Jul, 2023
- Subject
- Communication, Networking and Broadcast Technologies
Components, Circuits, Devices and Systems
Computing and Processing
Robotics and Control Systems
Brain
Wavelet domain
Databases
Magnetic resonance imaging
Software systems
Discrete wavelet transforms
Convolutional neural networks
Convolution neural network
Wavelet transform
Deep learning
Machine learning
Multiclass classification
- Language
The human brain is considered to be the most important organ in the body. Since the causes of brain cancer is still unknown, early detection is required for proper treatment. Magnetic Resonance Imaging (MRI) is an imaging technology used to depict inside structure of human body in details. This research study makes a contribution to the development of an image-based classification system as well as to the detection of brain cancer. Texture-based energy descriptors are retrieved using Discrete Wavelet Transform (DWT)-based sparse representation systems. A Convolutional Neural Network (CNN) in wavelet domain is used for further processing in order to improve the classification between normal and abnormal classes of MRI brain images. The performance of the proposed system is assessed by utilizing the standard collection of brain tumor images available in the REMBRANDT database. Results prove that the proposed system provides 98% accurate results for brain cancer classification.