Using Deep Learning Systems for Imaging Methods for Recognising Brain Tumors
Monjurul Islam Sumon, Md Tanvir Chowdhury, Syed Salman Hayat, Mahir Tagwar, Md Habibur Rahman, Md Sabbir Hossain, Mahamudul Hasan, Taskeed Jabid, Maheen Islam
Abstract
Uncontrolled and abnormal cell proliferation is the cause of a brain tumor, a type of cancer. The healthcare industry has benefited from recent advancements in deep learning by using diagnostic imaging to identify various disorders. Task If you want to learn about images or identify images, CNN is the machine learning algorithm you should use. In a similar vein, we have used the convolutional neural network (CNN) method, Data Augmentation, and Image Processing to determine whether or not images from brain MRI scans show cancer. Using the transfer learning method, we compared the performance of our shredded CNN model to that of the already trained Xception, which ResNet-50, and Inception-v3 models. Our model achieved a perfect score of 100% on a sizable dataset, demonstrating its high accuracy and low complexity, while Xception, ResNet-50, and Inception-V3 only managed to achieve 99.3%, 98.4%, and 99.4% accuracy, respectively. In comparison to other pre-trained models, our model only requires a little amount of processing resources and achieves substantially higher accuracy levels.