PERFORMANCE EVALUATION OF 3D VOLUMETRIC BRAIN TUMOR MRI IMAGEG SEGMENTATION

3D volumetric brain tumor segmentation presents a significant challenge in medical image analysis.The primary objective of brain tumor segmentation is to delineate the boundaries of the tumor within medical images accurately.In medical image applications, Convolutional Neural Networks (CNNs) have demonstrated significant progress, including semantic segmentation, classification, and object detection.To address complexities in detecting and segmenting brain tumor tissues in brain regions, deep learning has been specified as a promising solution.Early detection and classification of brain tumors are essential for effective treatment planning and patient prognosis.Deep learning techniques play a crucial role in detecting and diagnosing brain tumors using Magnetic Resonance Imaging (MRI).In various biomedical segmentation applications, the 3D U-Net architecture has presented effective performance.This paper proposes the use of a 3D-UNet model for semantic segmentation to analyze 3D volumetric brain tumors in MRI images.The proposed paper describes the trains of the 3D U-Net model with different hyperparameters.Additionally, tumor depth within 3D volumetric brain tumor MRI images is explored in this paper using the 3D U-Net.