Brain Tumor Detection in MRI Images Using Image Processing Techniques

Abstract

Biomedical technology plays an important role in the diagnosis and treatment of life-threatening diseases. Recently, one of the most common deadly diseases is brain tumors. The process of treating brain tumors depends on the experience and knowledge of physicians. For this reason, an automated tumor detection system is extremely important for radiologists and doctors to assist in the detection of brain tumors. Accurate analysis of magnetic resonance imaging (MRI) scans is needed to detect brain tumors and this paper proposes an effective method to predicting brain tumors by using convolution neural networks (CNNs). Methods include data augmentation and image pre-processing techniques that cut out the dark edges from images by finding extreme contours. The images are then normalized for a suitable scale. An Adaptive Moment Estimation (Adam) Optimizer has been added to expedite the training process in our network. MRI image dataset (own created Kaggle dataset) is used to train and evaluate the model to get maximized accuracy and the F1 score is calculated to evaluate the performance of the maximized model.