Brain tumor detection using HyGSNet and feature extraction with DWT-based GDP

A life-threatening condition that impacts neurological function is brain tumors, which can lead to psychiatric complications such as depression and panic attacks. Timely and accurate detection, followed by appropriate treatment, is essential to improve the quality of life. Quick and early recognition of brain tumors significantly enhances treatment outcomes and promotes effective healing. In this context, medical image processing plays a critical role in assisting clinicians to detect and classify brain abnormalities. However, the manual process is time-consuming and heavily reliant on the expertise of physicians. Therefore, an intelligent system for brain tumor detection is essential to support clinical decision-making. This research presented a Hybrid Google SpinalNet (HyGSNet) to automatically detect brain tumors from Magnetic resonance imaging (MRI) images. Here, the proposed HyGSNet model is the hybridization of GoogleNet and SpinalNet. Initially, the Adaptive Wiener filter is used for pre-processing the input image, and the UNeXt is used for the segmentation of the filtered image. Then, the image augmentation process is followed by feature extraction to extract the essential features. Finally, the extracted features are passed to the HyGSNet for detecting brain tumor. Here, the performance of HyGSNet is evaluated with various evaluation metrics. The HyGSNet approach recorded high performance with specificity of 93%, accuracy of 93%, and sensitivity of 93.7%. The experimental results demonstrate that the proposed approach achieves consistently high performance across key evaluation metrics, indicating its robustness and reliability for brain tumor detection.