U-Net and Its Variants for Medical Image Segmentation: A Review of Theory and Applications

Nahian Siddique,Vijay Devabhaktuni,Colin P Elkin,Sidike Paheding

doi:10.1109/access.2021.3086020

Nahian Siddique, Vijay Devabhaktuni + Show 2 more

Open Access

https://doi.org/10.1109/access.2021.3086020

Copy DOI

Abstract

U-net is an image segmentation technique developed primarily for image segmentation tasks. These traits provide U-net with a high utility within the medical imaging community and have resulted in extensive adoption of U-net as the primary tool for segmentation tasks in medical imaging. The success of U-net is evident in its widespread use in nearly all major image modalities, from CT scans and MRI to X-rays and microscopy. Furthermore, while U-net is largely a segmentation tool, there have been instances of the use of U-net in other applications. Given that U-net's potential is still increasing, this narrative literature review examines the numerous developments and breakthroughs in the U-net architecture and provides observations on recent trends. We also discuss the many innovations that have advanced in deep learning and discuss how these tools facilitate U-net. In addition, we review the different image modalities and application areas that have been enhanced by U-net.

Highlights

Thanks to recent advances in deep learning in computer vision within the past decade, deep learning has been increasingly utilized in the analysis of medical images
One such technique that will be discussed in this literature review will be the U-net, a deep learning technique widely adopted within the medical imaging community
We have found papers in which deep residual U-nets have been used to great effect in many biomedical imaging applications such as nuclei segmentation [52], [53], brain tissue quantification [41], brain structure mapping [54], retinal vessel segmentation [55], breast cancer [56], liver cancer [23], [57], prostate cancer [58], endoscopy [59], melanoma [59], osteosarcoma [60], bone structure analysis [61], and cardiac structure analysis [58], [62]

Summary

Introduction

Thanks to recent advances in deep learning in computer vision within the past decade, deep learning has been increasingly utilized in the analysis of medical images. There have been many breakthrough techniques over the years to overcome these various challenges, and new research is continuously leading to the development of more novel and innovative methods. One such technique that will be discussed in this literature review will be the U-net, a deep learning technique widely adopted within the medical imaging community. U-net is a neural network architecture designed primarily for image segmentation [1]. The first path is the contracting path, known as the encoder or the analysis path, which is similar to a regular convolution network and provides classification information. The second is an expansion path, known as the decoder or the synthesis path, The associate editor coordinating the review of this manuscript and approving it for publication was Junhua Li

Objectives

Findings

Discussion

Conclusion