Hephzibah R, Anandharaj HC, Kowsalya G, Jayanthi R, Chandy DA. Review on Deep Learning Methodologies in Medical Image Restoration and Segmentation.
Curr Med Imaging 2022;
19:844-854. [PMID:
35392788 DOI:
10.2174/1573405618666220407112825]
[Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/16/2021] [Revised: 01/16/2022] [Accepted: 01/25/2022] [Indexed: 11/22/2022]
Abstract
This paper presents a comprehensive review of two major image processing tasks such as restoration and segmentation in the medical field on deep learning perspective. These processes are essential as restoration helps in the removal of noise and segmentation helps in extracting the particular region of interest of an image which is essential for accurate diagnosis and treatment. This paper mainly focuses on deep learning techniques as it plays a prominent role over other conventional techniques in handling a large number of datasets in the medical field and also provides accurate results. In this paper, we reviewed the application of different convolutional neural network architectures in the restoration and segmentation processes. Based on the results in the case of image restoration, TLR-CNN and Stat-CNN are promising in achieving better PSNR, noise suppression, artifact suppression and improves the overall image quality. For segmentation process, LCP net achieves the Dice score as 98.12% and sensitivity as 98.95% in the cell contour segmentation;3D FCNN model is found to be the best method for segmentation of brain tumors. This review work shows that deep learning methodologies can be a better alternative for medical image restoration and segmentation tasks as the data size is an important concern as on today.
Collapse