Brain tumor recognition using an integrated bat algorithm with a convolutional neural network approach

Abstract

Tumors are uncontrolled growth of nerve cells that can lead to cancer.The ability to separate malignancies in the brain is critical for better diagnosis. As a result, there are a few research initiatives aimed at improving patient care. The manual method takes time and is only available in a few medical facilities. Treatment choices for brain cancer differ depending on the specific, shape, and position of the tumor, as well as your general health and preferences.Based on MRI input images, the system can automatically recognize a type of brain tumor. A Convolutional neural network and Bat algorithm are used in the proposed method to detect brain tumors in MRI images (B–CNN).To eliminate the noise the data is first pre-processed. To extract features from MRI brain pictures, the 2-D Gabor filter is utilized. For better accuracy, feature selection is based on the Bat algorithm.The data for this research were obtained from Nanfang Hospital and Tianjing Medical University's General Hospital.The proposed BCNN method results in a 99.5% accuracy rate when compared to the existing system.

Keywords

• Brain tumor
• Convolutional neural network
• Bat algorithm
• 2-D gabor filter
• MRI Images
• Classification