Abstract:
Classification of brain tumors is one of the most crucial jobs in medical imaging, and deep
learning models have shown promising outcomes when it comes to automation. We
provide a thorough analysis of three deep learning models for brain tumor classification in
this research utilizing a dataset of various types of MRI images of brain tumors.
Convolutional Neural Networks (CNNs), VGG16, and InceptionV3 are the names of these
proprietary models. Classifying brain tumors using a huge dataset of magnetic resonance
imaging (MRI) pictures is the aim of this effort. No ionizing radiation is used during an
MRI, making it a safer and more thorough way to learn about the anatomy. A convolutional
neural network (CNN) is trained on several datasets, such as images of benign tumors,
meningiomas, gliomas, and pituitaries, in order to develop a robust prediction model. The
model's goal is to evaluate MRI images automatically and distinguish between brain areas
that are tumor-filled and those that are normal. If this effort be successful, it will enable
prompt intervention and customized treatment plans by enabling early, non-invasive
identification. By providing a trustworthy method for categorizing brain tumors, this work
enhances medical imaging.
