Image segmentation is indispensable in analyzing medical images. By splitting the images into separate segments or regions different anatomical structures or potential pathologies can be identified. A promising approach to segmentation is the use of hybrid techniques, which combine multiple methods to improve accuracy and robustness. These hybrid techniques often incorporate image processing methods to further enhance the segmentation results. Recently, various methods like DenseNet and U-Net have become increasingly popular in medical image segmentation because of their capacity to identify the complex patterns and features from the images. The combination of DenseNet and U-Net has resulted in a novel algorithm that demonstrates superior accuracy compared to other existing algorithms.