基于U-Net网络的医学图像分割研究综述OACSTPCD

With the rapid development of deep learning technology in recent years,convolutional neural network(CNN)has become an important support framework for semantic segmentation and is widely used in a variety of target detection and segmentation tasks.In medical image segmentation tasks,U-Net network has become a hot research topic in this field with its excellent segmentation performance and expandable network structure.Nowadays,many scholars have improved U-Net in terms of the structure of the network to optimize the network performance and improve the segmentation accuracy.The study first introduces the classical improved model based on U-Net by analyzing the relevant literature.Then,six U-Net improvement mechanisms are described:attention mechanism,inception module,residual structure,dilated mechanism,dense connection structure and integrated network structure.Common evaluation metrics and unstructured improvement schemes for medical image segmentation are then presented.These unstructured improvement methods include four aspects of data enhancement,optimizers,activation functions,and loss functions.After that,improved models in four major medical image segmentation areas,namely,pulmonary nodules,retinal vessels,skin diseases and intracranial tumors,are listed and analyzed.Finally,the future development of U-Net network is prospected to provide ideas for related research.