Medical image segmentation is a critical task in the field of healthcare. It helps in identifying and visualizing anatomical structures, in various imaging modalities (e.g., CT, MRI, ultrasound), which aids healthcare professionals to interpret medical images and diagnosing diseases more accurately. Today, Deep Learning-Based Segmentation approaches are the first choice to solve many problems in medical image segmentation compared with traditional segmentation techniques like; Thresholding, Region-Based method and Edge-Based method. This paper provides a review on deep learning models for semantic segmentation applied to medical images, it explores various neural network architectures, including convolutional neural networks (CNNs), fully convolutional networks (FCNs), U-Net, and more recent innovations such SegNet and DeepLabv3+, the paper also discusses different evaluation metrics used to evaluate segmentation performance, including accuracy, Dice coefficient, intersection over union (IoU) and F1-Score.