Classifying and Localizing Abnormalities in Brain MRI Using Channel Attention Based Semi-Bayesian Ensemble Voting Mechanism and Convolutional Auto-Encoder

Abstract

Adults who have brain tumors face a serious and potentially fatal challenge since the tumors rapidly growing malignant cells can seriously impair their physiological ability. In clinical practice, imaging techniques such as MRI, PET, and CT scans are widely used to determine the size, kind, and location of tumors. The goal of this research is to create a Computer-Aided Diagnostic (CAD) system that can automatically segment and classify brain tumors utilizing T1W-CE Magnetic Resonance Imaging (MRI) of the brain. The CAD system will involve two primary tasks: tumor classification, which determines the type of tumor depicted in the image, and tumor segmentation, which involves accurately determine the tumor region from the surrounding healthy tissue. By automating these processes, the proposed system aims to enhance the accuracy and effectiveness of brain tumor diagnosis and treatment planning. The classification of brain tumors into multiple classes is recognized as a complex challenge within the field of medical imaging. This research article proposes a model named VS-BEAM that can be used efficiently for clinical decision-making. The proposed VS-BEAM (Voting Based Semi-Supervised Bayesian Ensemble Attention Mechanism) model has been examined for a brain tumor’s multi-class classification. The VS-BEAM model achieved the highest level of accuracy possible. The proposed work achieves maximum sensitivity, specificity, and diagnostic accuracy compared to existing models using T1W-CE MRI images. A convolutional autoencoder is utilized for extracting tumors from MRI images. The accuracy obtained from testing data of 264 brain tumors was 98.91%, indicating that the method is effective and can be used in the clinical context to assist in detecting larger or even smaller tumors.

Keywords

• Magnetic resonance imaging
• medical imaging
• squeeze and excitation networks
• Bayesian learning
• ensemble learning
• convolutional autoencoder