Automated Prediction of Glioma Using Deep Learning and Transfer Learning Models

New advances in deep learning have made it possible to develop better disease prediction and classification tools for the healthcare industry. Glioma is one of the most prevalent forms of brain cancer and is associated with increased risks of brain cancer. It is critical to ensure early and accurate diagnosis of gliomas to help healthcare professionals create effective strategies for treatment. Transfer learning has proven to be an effective technique for developing automatic glioma prediction tools based on medical imaging. MRI images used in the process are subjected to pre-processing, feature extraction, and classification procedures to enable accurate identification of tumors. In this research, the EfficientNetB3 transfer learning model is used for glioma classification, and YOLOv8 deep learning models are applied for detecting and localizing the tumor. Moreover, CNN models are used to increase the predictability of the developed system. Despite the fact that current literature primarily discusses the use of traditional machine learning methods in predicting brain tumors, the suggested research relies on deep learning and transfer learning models to improve the quality of glioma prediction. The results of the experiments conducted prove the efficiency of the suggested approach in glioma identification based on MRI images.