Brain Tumor Diagnosis Using CNN

Brain tumors are deadly neurological conditions caused by uncontrolled cell growth in the brain or skull. This disease's mortality rate continues to rise. Early detection of malignant tumors is essential for treating patients and improving survival rates. The evaluation and treatment of brain tumors rely heavily on the accurate categorization of the tumors. To spot malignant growths in the brain, neuroradiologists employ a variety of imaging methods. However, Magnetic Resonance Imaging is widely employed because it has a high picture quality as well as its absence of reliance on ionizing radiation. When it comes to classification and segmentation difficulties, among the machine learning subfields is deep learning (DL), which has lately demonstrated impressive results. Utilizing techniques of DL in the context of a big dataset medical field has shown to be quite advantageous, opening up completely new pathways and opportunities for the development of novel methods for the detection of illnesses. During this investigation, we are going to discuss a deep learning model that was created with the help of a CNN architecture to identify and categorize various forms of brain tumors from data found in the open-source dataset. The identification of brain tumors is accomplished with the help of a simple convolutional model, and the categorization of the primary kinds of brain tumors is carried out with the assistance of a transfer learning model known as ResNet50.