BrainVisionNet: A Deep Learning-based Approach to Evaluate the Potential of Microwave Imaging for Classification of Brain Tumors

Brain tumor diagnosis and categorization are essential for patient care and therapy planning. Microwave imaging has emerged as a viable non-invasive and cost-effective technology for brain imaging. Compared to current imaging methods like X-rays and CT scans, which employ ionizing radiation that can harm healthy tissues, and MRI, which requires a large setup of power-hungry equipment, microwave imaging is more favorable. This research proposes a newly developed model called BrainVisionNet based on a convolutional neural network to identify brain tumors from microwave images. Several evaluation metrics, including accuracy, precision, recall, and F1-Score, have been employed to assess the performance of this model. The suggested method of brain tumor classification using microwave images demonstrates superior performance than the existing methods, obtaining an overall accuracy, f1-score, precision, and recall of 99.20%, 99%, 99.3%, and 99.3% respectively, according to our testing results. This study illustrates the potential of microwave imaging, for accurate and efficient categorization of brain tumors utilizing BrainVisionNet. The findings imply that BrainVisionNet can be a useful tool for radiologists and physicians in the precise and fast diagnosis of brain tumors from microwave images.