Brain Tumour Disease Detection in MRI Images using Xception Model

The main aim of this study is to examine the utilization of the Xception transfer learning model for the crucial task of detecting brain tumor diseases in medical images, with a specific focus on improving mortality rates. This study highlights the urgent requirement for automated and precise techniques for detecting brain tumors in medical imaging. The present study proposed a novel Xception model architecture to classify brain tumor illness images into two categories. The study included a dataset consisting of 253 photos. The model completed training for a total of 40 epochs, employing a batch size of 8 and utilizing the default learning rate. The model training procedure involves optimizing it using a combination of Adam and Adamax optimizers. The findings illustrate the model's competence, achieving a noteworthy level of accuracy at 96%. The results of this discoveries are of considerable importance, as they demonstrate the potential of the Xception model in furthering the field of medical diagnostics. The attainment of a high level of accuracy implies enhanced capabilities in the early diagnosis and prompt intervention of brain tumor disorders, potentially leading to significant consequences for patient outcomes and healthcare practices.