Streamlining Brain Tumor Classification with Custom Transfer Learning in MRI Images
This work addresses the need for efficient brain tumor diagnosis in clinical settings, but it is incremental as it builds on existing pre-trained architectures.
The study tackled the problem of high computational complexity in brain tumor classification from MRI images by proposing a custom lightweight transfer learning model based on VGG-19 with additional hidden layers, achieving a classification accuracy of 96.42%.
Brain tumors are increasingly prevalent, characterized by the uncontrolled spread of aberrant tissues in the brain, with almost 700,000 new cases diagnosed globally each year. Magnetic Resonance Imaging (MRI) is commonly used for the diagnosis of brain tumors and accurate classification is a critical clinical procedure. In this study, we propose an efficient solution for classifying brain tumors from MRI images using custom transfer learning networks. While several researchers have employed various pre-trained architectures such as RESNET-50, ALEXNET, VGG-16, and VGG-19, these methods often suffer from high computational complexity. To address this issue, we present a custom and lightweight model using a Convolutional Neural Network-based pre-trained architecture with reduced complexity. Specifically, we employ the VGG-19 architecture with additional hidden layers, which reduces the complexity of the base architecture but improves computational efficiency. The objective is to achieve high classification accuracy using a novel approach. Finally, the result demonstrates a classification accuracy of 96.42%.