Abstract

An aberrant cell growth in the brain is called a brain tumor. These growths may be malignant (cancerous) or benign (non-cancerous). Benign tumors are generally slower-growing and less likely to invade nearby tissues, while malignant tumors can grow more rapidly and may spread to other parts of the brain or the body. Delay in the detection of a brain tumor may result in the tumor growing larger and potentially becoming more challenging to treat. Hence, early detection and treatment of a brain tumor are crucial for treatment options, preventing further damage, improved prognosis, and for preventing complications. This paper aims to improve the performance of brain tumor detection using Convolution Neural Network (CNN) with accuracy 98%. Transfer Learning technique is applied on CNN after taking a pre-trained model on a large dataset and fine-tuning it.