Early detection of brain tumors: Harnessing the power of GRU networks and hybrid dwarf mongoose optimization algorithm

Abstract

Brain tumor detection is a challenging problem that requires accurate and robust methods to identify and locate the abnormal regions in the brain images. MRI is the most commonly used imaging modality for brain tumor diagnosis, as it can provide high-resolution and contrast images of the brain tissues. However, manual analysis of MRI images by human experts is time-consuming, subjective, and prone to errors. Therefore, there is a need for automated and intelligent methods that can analyze the MRI images and detect brain tumors effectively and efficiently. In this paper, we propose a novel machine learning method that combines the advantages of the Gated Recurrent Unit (GRU) networks and the Enhanced Hybrid Dwarf Mongoose Optimization (EHDMO) algorithm for brain tumor detection. The GRU networks are a type of Recurrent Neural Network (RNN) that can process sequential data, such as natural language or time series. We employ the EHDMO algorithm to fine-tune the parameters of the GRU networks, such as the weight matrices and bias vectors for each gate and the candidate's hidden state, along with the number of hidden units in the network. The proposed method is applied to the brain tumor detection problem using the “Brain-Tumor-Progression” dataset. Results show that the proposed method achieves a sensitivity of 0.98, a specificity of 0.97, a PPV of 0.98, an NPV of 0.98, and an accuracy of 0.95. These results indicate that the proposed method can accurately and robustly detect brain tumors from MRI images. The method also is compared with some of the most recent methods, such as BrainMRNet, VGG19, ASSO, CNN/POA, and YOLOv2. The proposed method outperforms these methods in terms of sensitivity, specificity, PPV, NPV, and accuracy, demonstrating its effectiveness and efficiency.