•  
  •  
 

Article Type

Original Study

Abstract

Brain cancer is considered one of the most dangerous types that must be treated as soon as possible. Therefore, it is necessary to detect brain cancer in its early stages to enhance the diagnosis of the condition. This study proposes to combine the bat algorithm (BA) with deep learning methods to identify and classify brain tumors in medical images accurately.

In this paper, the bat algorithm was used to select distinctive features from the input brain images. The Bat Algorithm (BA) is a metaheuristic optimization algorithm that mimics the echolocation behavior of bats. It efficiently explores the feature space to discover the most suitable features for analysis, which is useful and employs these features to train an exceptionally advanced artificial neural network (ANN), the Convolutional Neural Network (CNN), that improves the classification of cancerous brain tissue from normal tissue. Furthermore, a deep learning model can accurately depict complex nonlinear patterns within data, enhancing diagnostic capabilities.

A series of tests were performed using readily available MRI and CT scans obtained from the public domain. Combining the bat algorithm with a deep learning model outperforms previously used techniques for feature extraction and classification of brain tumors, leading to more accurate detection. We have noticed an improvement in terms of accuracy, sensitivity, and selectivity, with accuracy reaching

Keywords

Mean filter, Bat algorithm, Artificial neural network (ANN), Convolutional neural network (CNN)

Download

Share

COinS