MRI Based Brain Tumor classification Using a Fine-Tuned EfficientNetB3 Transfer Learning Model

In this paper, the formidable task of classifying brain tumors in MRI imaging was addressed by employing an extensive compilation of brain tumor images. We showed that precisely adjusting a cutting-edge efficientNetB3 model using transfer learning considerably enhanced its performance in classifying brain cancers. the experimentation employs a hybrid dataset from the Kaggle, which is a combination of two different datasets (Kaggle brain-tumor-classification-mri and Kaggle brain-tumor). The proposed fine-tuned efficientNetB3 transfer learning model showed promising results by classifying multi-class brain tumors with a testing accuracy of 98.13%.