Image Analysis for MRI-Based Brain Tumor Classification Using Deep Learning

Tumors are cells that grow abnormally and uncontrollably, whereas brain tumors growing in or near the brain. It is estimated 23,890 adults (13,590 males 10,300 females) United States 3,540 children under age of 15 would be diagnosed with a tumor. Meanwhile, there over 250 cases Indonesia patients afflicted tumors, both infants. The doctor medical personnel usually conducted radiological test commonly performed using magnetic resonance image (MRI) to identify From several studies, each researcher claims results their proposed method can detect high accuracy; however, still flaws methods. This paper will discuss classification MRI-based deep learning transfer learning. Transfer allows for various domains, functions, distributions used training research. research public dataset. dataset comprises 253 images, divided into 98 tumor-free images 155 tumor images. Residual Network (ResNet), Neural Architecture Search (NASNet), Xception, DenseNet, Visual Geometry Group (VGG) techniques use this paper. got show ResNet50 model gets 96% accuracy, VGG16 accuracy. obtained indicate handle