The orientation of medical images is important in the diagnosis of diseases, making proper segmentation of images an essential aspect. U-Net and its modifications have been widely used in medical image segmentation among different architectures. However, they also present certain disadvantages concerning the analysis of such complex structures as cell morphologies or multi-scale structures. In this work, the N-Net model has been applied, which is based on the U-Net model and adds two encoders, Squeeze-and-Excitation and full-scale skip connectors for image segmentation. Results show that the N- Net outperforms the previous methods in terms of segmentation accuracy on the LIDC-IDRI dataset, making a step forward in the field of medical image processing.