DEEP LEARNING BASED BRAIN TUMOR CLASSIFICATION USING MAGNETIC RESONANCE IMAGING
Unknown authors
Abstract
Brain tumor means the aggregation of abnormal cells in some tissues of the brain. Brain tumor can be cancerous or noncancerous. The most common types of brain tumors are Glioma, Meningioma and Pituitary tumor. Early detection of tumor cells plays a major role in treatment and recovery of patient. Diagnosing a brain tumor usually undergoes a very complicated and time consuming process. The MRI images of various patients at various stages can be used for the detection of tumors. There are various types of feature extraction and classification methods which are used for detection of brain tumor from MRI images. Convolutional Neural Network image classification algorithm helps in detecting the tumor at early stage with high accuracy. We proposed a Recurrent Neural Network architecture for detection of tumor cells which gives accuracy of about 90%. A recurrent neural network (RNN) is a type of artificial neural networks in that connections between nodes form a directed graph along a temporal sequence.