Enhanced Brain Tumor Detection: A Novel CNN Approach Optimized by the Crow Search Algorithm

Articles in Press

Document Type : Research Paper

Authors

1 Institute of Artificial Intelligence, Social and Advanced Technologies, Ya.C., Islamic Azad University, Yazd, Iran.

2 Department of Computer Engineering, Meybod University, Meybod, Iran

Abstract

Medical imaging serves as a vital tool for diagnosing various diseases. These images enable doctors to assess conditions with greater accuracy. However, the manual identification and analysis of large amounts of Magnetic Resonance Imaging (MRI) data is challenging and time-consuming. Consequently, there is a critical need for a reliable deep learning (DL) model that can accurately detect brain tumors. Deep learning techniques, such as Convolutional Neural Networks (CNN), have proven to be very effective in identifying brain tumors. Nevertheless, despite their effectiveness, CNNs face several challenges when used for brain tumor detection based on medical imaging, including inadequate extraction of image texture features and reliance on a single classifier in the fully connected layer, which complicates the segmentation process in CNN architecture. In this context, the necessity of employing metaheuristic algorithms arises, which aim to find the optimal solution among existing alternatives. In this study, a novel approach for classifying and segmenting brain tumors using CNN and the Crow Search Algorithm is presented. This algorithm enhances CNN parameters, helping to achieve more accurate results with the best learning rate. Simulation results demonstrate the superiority of this method over the Harris Hawk optimization algorithm.

Keywords

Main Subjects

Journal of Modeling in Engineering

Articles in Press, Accepted Manuscript
Available Online from 17 June 2026

Files

History

  • Receive Date: 05 August 2025
  • Revise Date: 13 March 2026
  • Accept Date: 16 June 2026

Share

How to cite

Statistics