Brain tumors are the most common and aggressive disease, with a noticeably short life expectancy in their most severe form. Thus, treatment planning is an important stage in improving patients' quality of life. In general, image techniques such as computed tomography (CT), magnetic resonance imaging (MRI), and ultrasound images are used to evaluate tumors in the brain, lung, liver, breast, prostate, and so on. MRI images are used in this work to diagnose brain tumors. However, the massive amount of data generated by an MRI scan makes manual classification of tumor vs non-tumor in each time impossible.
The automatic classification of brain tumors is a challenging task due to the large spatial and structural variability of the surrounding region of the brain tumor. Automatic brain tumor detection using Convolutional Neural Networks (CNN) classification is proposed in this work. Small kernels are used to perform the deeper architecture design. The neuron's weight is given as small. When compared to all other state-of-the-art methods, experimental results show that CNN archives rate of 97.5 percent accuracy with low complexity.
In this project,the model will take MRI images of the patient and determine whether there is a tumor in the brain. Using publicly available Kaggle Brain MRI Images for Brain Tumor Detection, the model will be trained and predict the images using the gui.