Abstract:
Determining the amount of the tumor may be a particularly difficult process among all of them; the brain tumor is the toughest one. Brain tumors account for 85% to 90% of all primary central nervous system (CNS) tumors. One of the discrete methods that emerged as a first-line, radiation-free method of diagnosing brain tumors is magnetic resonance imaging (MRI). When it comes to picture identification, deep learning has made significant progress. Work on transitioning from convolutional neural networks (CNN) and visual geometry group (VGG-16) alternative autoencoders has found countless applications in the realm of medical image analysis, propelling it forward quickly. In radiology, the skilled physician externally evaluated clinical images for the identification, depiction, and observation of illnesses. In this study, machine learning and classifications using convolution neural networks (CNNs) are offered as a technique for automatically detecting brain tumors. Small kernels are responsible for designing the deeper architecture. There is a minor weight assigned for the neuron. In comparison to all other methods, it has been found that VGG-16 achieves a high rate of accuracy with a minimal level of complexity. The increased accuracy will facilitate effective medical care and other institutions that work with brain tumors or any other related diseases.
Keywords: Brain Tumor, MRI, CNN, VGG-16.
