Detection of Brain Tumors Using Magnetic Resonance Images through the Application of an Innovative Convolution Neural Network Model

Abstract

According to a report released by the WHO in February 2018, the mortality rate for people with brain or central nervous system cancer is highest in Asia. It is important that cancer screenings are conducted earlier to prevent these deaths. Due to the complexity of brain cancer diagnosis, it is very important that the development of effective and non-invasive tools for analyzing and predicting the grade of the disease is carried out. Currently, there are various imaging modalities that can be used to detect brain tumors, such as CT, MRI, and X-rays. Deep Learning is a type of artificial intelligence that imitates the brain's work. It can learn to recognize and interpret the voice, make decisions, and translate languages. It can also detect artifacts in data, and without human intervention, it can understand from unorganized information. A Convolutional Neural Network is a type of deep learning that is commonly used in optical representation analysis. Currently, there are systems that can detect brain tumors using small datasets. However, they only use image processing techniques and require a lot of computational resources. A new system that combines the three components of deep learning, namely image preprocessing, augmentation, and applying, is currently under development.