A comparison of MobileNetV2, VGG16, and ResNet50 for classifying brain tumors

A terrible type of cancer, brain tumors pose a threat to the lives of more than 500,000 people globally. Researchers are focusing on brain tumor detection because early detection is essential to cure this disease. Currently, the primary method for identifying brain tumors is the analysis of magnetic resonance imaging (MRI) images of the human brain. Deep learning algorithms have the potential to greatly improve detection efficiency and accuracy. The aim of this investigation was to assess the efficacy of three Convolutional Neural Network (CNN) models—ResNet50, VGG16, and MobileNetV2—in the classification task of MRI images of brain tumors. 3,265 MRI pictures of the human brain make up the dataset. Following training, the models’ respective prediction accuracies were 92.05%, 91.13%, and 85.32%. The comparison showed that MobileNetV2 did the poorest and ResNet50 did the best. These results demonstrate ResNet’s potential as a very successful model for classifying brain tumors, providing a viable instrument to support early diagnosis. To improve clinical applicability and diagnostic accuracy, future research could investigate more optimization and practical application.