Classification of Brain Tumor Types on Magnetic Resonance Images Using Hybrid Deep Learning Approach with Radial Basis Function Neural Network | Chapter 05 | Recent Advances in Mathematical Research and Computer Science Vol. 7

 In general, classification or segmentation refers to the division of an image into smaller sections in order to identify or pinpoint the abnormality region. Even though it remains a difficult task in medical applications due to the complicated structure of brain tumours, due to opposing image, local observations of an image, noise image, and non uniform texture of the images, and so on. Although there are numerous image segmentation techniques accessible, it is still necessary to introduce efficient and quick medical picture segmentation approaches. This research paper presents an effective picture classification method based on K-means clustering and a K-means clustering algorithm. The following are the primary contributions of this study paper: The first machine learning classification model for brain tumours was built on a Hybrid PSO-WCA (Particle Swarm Optimization-Water Cycle Algorithm) based Radial Basis Function Neural Network (RBFNN). Second, the GLCM (Gray Level Co-occurrence Matrix) approach was utilised to extract features. Finally, the performance of the clustering classification was demonstrated, with malignant and benign tumours being identified using Fast fuzzy c-means, KNN (Nearest Neighbor) method, Fuzzy c means algorithm, and K-Means algorithm with features as input for visual localization. Finally, the extracted tumour images from the proposed hybrid model PSO-WCA-RBFNN are segmented to produce 99.62 percent accurate representations of brain tumours. The suggested technique was also compared to various current segmentation algorithm models such as PSO-RBFNN, WCA-RBFNN, and LMS-RBFNN. The results suggest that the proposed method has a higher accuracy rate. 255 MRI brain scans from Harvard Medical School are used in the investigation.

Author(S) Details

T. Gopi Krishna
Department of CSE, SoEEC, Adama Science and Technology University, Ethiopia.

Satyasis Mishra
Department of ECE, SoEEC, Adama Science and Technology University, Ethiopia.

Sunita Satapathy
Department of Zoology, School of Applied Science, Centurion University of Technology & Management, Odisha, India.

K. V. N. Sunitha
Department of CSE, BVRIT Women’s Engineering College, Hyderabad, India

Mohamed A. Abdelhadi
Department of Information Technology, Tripoli University, Libya

View Book:- https://stm.bookpi.org/RAMRCS-V7/article/view/5490