A Comparative Analysis of Pretrained Models for Brain Tumaor Classification and Their Optimization Using Optuna

In our current times, with the alarming spread of brain tumors of various types and causes, the development and improvement of deep models to expedite and simplify the process of detecting and classifying brain tumors is of utmost importance. Therefore, in this study, we conduct a comparative analysis of three pre-trained deep learning models VGG19, EfficientNetB3, and ResNet50 with the Brain Tumor Magnetic Resonance Imaging dataset (BTMRI) from Kaggle. The models were evaluated based on their validation and test accuracy in classifying brain tumor images. Initially, the models trained on the dataset to evaluate their performances, and then the performance of these models was enhanced using the Optuna hyperparameters optimization library, to tune three main hyperparameters: learning rate, dropout rate, and the number of epochs, due to this the models achieved higher validation and testing accuracy rates of 95.42% and 94.51% for VGG19, 99.54% and 99.39% for EfficientNetB3, and 98.78% and 98.93% for ResNet50, respectively. Overall This study highlights the effectiveness of deep learning models in medical image classification and the impact of hyperparameter optimization in improving model performance.