Medical Image Super-Resolution Reconstruction Algorithms on Deep Learning

DOI

10.2991/978-94-6463-370-2_41How to use a DOI?

Keywords

Medical image; Super-resolution reconstruction; Convolutional neural networks; MSID; NSST

Abstract

The human body’s structural details can be more clearly seen in high-resolution MRI and CT pictures, which can also aid in the early identification of disorders. However, due to the limits of imaging technologies, imaging surroundings, and human variables, clean high-resolution photographs are challenging to obtain. For super-resolution reconstruction of medical pictures, I propose a non-subsampled shearlet transform (NSST) and multi-scale information distillation (MSID) network in the current study namely NSST-MSID network. In order to thoroughly investigate the multiscale aspects of images and successfully restore low-resolution photos to high-resolution images, an MSID network that primarily comprises of several cascaded MSID blocks is first proposed. In addition, the super-resolution problem of medical images is characterized as a prediction problem of NSST coefficients, so that the MSID network maintains richer structural details than the spatial domain. This is because existing methods frequently predict high-resolution images in the spatial domain, making the output too smooth and texture details lost. The performance of the suggested strategy is then assessed using the well-known medical image dataset. In comparison to other outstanding methods currently in use, the experimental results demonstrate that the NSST-MSID network can achieve better peak signal-to-noise ratio (PSNR), structural similarity (SSIM), and root-mean-square error (RMSE) values while better preserving local texture details and global topology.

Copyright

© 2024 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Jinglin Yuan
PY  - 2024
DA  - 2024/02/14
TI  - Medical Image Super-Resolution Reconstruction Algorithms on Deep Learning
BT  - Proceedings of the 2023 International Conference on Data Science, Advanced Algorithm and Intelligent Computing (DAI 2023)
PB  - Atlantis Press
SP  - 389
EP  - 399
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6463-370-2_41
DO  - 10.2991/978-94-6463-370-2_41
ID  - Yuan2024
ER  -