Open Access

ARTICLE

A Novel Handcrafted with Deep Features Based Brain Tumor Diagnosis Model

Abdul Rahaman Wahab Sait^1,*, Mohamad Khairi Ishak²

1 Department of Documents and Archive, Center of Documents and Administrative Communication, King Faisal University, Al Hofuf, 31982, Al-Ahsa, Saudi Arabia
2 School of Electrical and Electronic Engineering, Engineering Campus, Universiti Sains Malaysia (USM), Nibong Tebal, 14300, Penang, Malaysia

* Corresponding Author: Abdul Rahaman Wahab Sait. Email:

Intelligent Automation & Soft Computing 2023, 35(2), 2057-2070. https://doi.org/10.32604/iasc.2023.029602

Received 07 March 2022; Accepted 14 April 2022; Issue published 19 July 2022

Abstract

In healthcare sector, image classification is one of the crucial problems that impact the quality output from image processing domain. The purpose of image classification is to categorize different healthcare images under various class labels which in turn helps in the detection and management of diseases. Magnetic Resonance Imaging (MRI) is one of the effective non-invasive strategies that generate a huge and distinct number of tissue contrasts in every imaging modality. This technique is commonly utilized by healthcare professionals for Brain Tumor (BT) diagnosis. With recent advancements in Machine Learning (ML) and Deep Learning (DL) models, it is possible to detect the tumor from images automatically, using a computer-aided design. The current study focuses on the design of automated Deep Learning-based BT Detection and Classification model using MRI images (DLBTDC-MRI). The proposed DLBTDC-MRI technique aims at detecting and classifying different stages of BT. The proposed DLBTDC-MRI technique involves median filtering technique to remove the noise and enhance the quality of MRI images. Besides, morphological operations-based image segmentation approach is also applied to determine the BT-affected regions in brain MRI image. Moreover, a fusion of handcrafted deep features using VGGNet is utilized to derive a valuable set of feature vectors. Finally, Artificial Fish Swarm Optimization (AFSO) with Artificial Neural Network (ANN) model is utilized as a classifier to decide the presence of BT. In order to assess the enhanced BT classification performance of the proposed model, a comprehensive set of simulations was performed on benchmark dataset and the results were validated under several measures.

---

Keywords

---