This study presents a novel U-Net-based framework for anatomy segmentation in MRI scans, utilizing a pre-trained ResNet50 backbone to enhance feature extraction. The framework is comprehensively evaluated across different anatomical planes, with optimization of both loss functions and image scales. The axial plane achieved the best performance, with a Dice score of 0.91 using the baseline model. By combining Dice loss and boundary loss in an optimal ratio, and setting the input image scale to 1.0, the model achieved an average Dice score of 0.92 across 10 diverse datasets, demonstrating its robustness. Furthermore, this model is integrated into MedVis Suite, an open-sourced educational platform designed to facilitate learning in computer vision and machine learning, offering students hands-on experience with medical imaging tasks.