4× Super‐resolution of unsupervised CT images based on GAN
Abstract Improving the resolution of computed tomography (CT) medical images can help doctors more accurately identify lesions, which is important in clinical diagnosis. In the absence of natural paired datasets of high resolution and low resolution image pairs, we abandoned the conventional Bicubic...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2023-06-01
|
| Series: | IET Image Processing |
| Subjects: | |
| Online Access: | https://doi.org/10.1049/ipr2.12797 |
| _version_ | 1797813808656285696 |
|---|---|
| author | Yunhe Li Lunqiang Chen Bo Li Huiyan Zhao |
| author_facet | Yunhe Li Lunqiang Chen Bo Li Huiyan Zhao |
| author_sort | Yunhe Li |
| collection | DOAJ |
| description | Abstract Improving the resolution of computed tomography (CT) medical images can help doctors more accurately identify lesions, which is important in clinical diagnosis. In the absence of natural paired datasets of high resolution and low resolution image pairs, we abandoned the conventional Bicubic method and innovatively used a dataset of images of a single resolution to create near‐natural high–low‐resolution image pairs by designing a deep learning network and utilizing noise injection. In addition, we propose a super‐resolution generative adversarial network called KerSRGAN which includes a super‐resolution generator, super‐resolution discriminator, and super‐resolution feature extractor to achieve a 4× super‐resolution of CT images. The results of an experimental evaluation show that KerSRGAN achieved superior performance compared to the state‐of‐the‐art methods in terms of a quantitative comparison of non‐reference image quality evaluation indicators on the generated 4× super‐resolution CT images. Moreover, in terms of an intuitive visual comparison, the images generated by the KerSRGAN method had more precise details and better perceptual quality. |
| first_indexed | 2024-03-13T07:58:11Z |
| format | Article |
| id | doaj.art-7bba0b39b23d4e05bdd3971fdcfa99f4 |
| institution | Directory Open Access Journal |
| issn | 1751-9659 1751-9667 |
| language | English |
| last_indexed | 2024-03-13T07:58:11Z |
| publishDate | 2023-06-01 |
| publisher | Wiley |
| record_format | Article |
| series | IET Image Processing |
| spelling | doaj.art-7bba0b39b23d4e05bdd3971fdcfa99f42023-06-02T03:06:38ZengWileyIET Image Processing1751-96591751-96672023-06-011782362237410.1049/ipr2.127974× Super‐resolution of unsupervised CT images based on GANYunhe Li0Lunqiang Chen1Bo Li2Huiyan Zhao3School of Electronic and Electrical Engineering Zhaoqing University Zhaoqing ChinaSchool of Electronic and Electrical Engineering Zhaoqing University Zhaoqing ChinaSchool of Electronic and Electrical Engineering Zhaoqing University Zhaoqing ChinaSchool of Electrical Engineering & Information Northeast Petroleum University Daqing ChinaAbstract Improving the resolution of computed tomography (CT) medical images can help doctors more accurately identify lesions, which is important in clinical diagnosis. In the absence of natural paired datasets of high resolution and low resolution image pairs, we abandoned the conventional Bicubic method and innovatively used a dataset of images of a single resolution to create near‐natural high–low‐resolution image pairs by designing a deep learning network and utilizing noise injection. In addition, we propose a super‐resolution generative adversarial network called KerSRGAN which includes a super‐resolution generator, super‐resolution discriminator, and super‐resolution feature extractor to achieve a 4× super‐resolution of CT images. The results of an experimental evaluation show that KerSRGAN achieved superior performance compared to the state‐of‐the‐art methods in terms of a quantitative comparison of non‐reference image quality evaluation indicators on the generated 4× super‐resolution CT images. Moreover, in terms of an intuitive visual comparison, the images generated by the KerSRGAN method had more precise details and better perceptual quality.https://doi.org/10.1049/ipr2.12797computed tomography imagesgenerative adversarial networksuper‐resolution |
| spellingShingle | Yunhe Li Lunqiang Chen Bo Li Huiyan Zhao 4× Super‐resolution of unsupervised CT images based on GAN IET Image Processing computed tomography images generative adversarial network super‐resolution |
| title | 4× Super‐resolution of unsupervised CT images based on GAN |
| title_full | 4× Super‐resolution of unsupervised CT images based on GAN |
| title_fullStr | 4× Super‐resolution of unsupervised CT images based on GAN |
| title_full_unstemmed | 4× Super‐resolution of unsupervised CT images based on GAN |
| title_short | 4× Super‐resolution of unsupervised CT images based on GAN |
| title_sort | 4 super resolution of unsupervised ct images based on gan |
| topic | computed tomography images generative adversarial network super‐resolution |
| url | https://doi.org/10.1049/ipr2.12797 |
| work_keys_str_mv | AT yunheli 4superresolutionofunsupervisedctimagesbasedongan AT lunqiangchen 4superresolutionofunsupervisedctimagesbasedongan AT boli 4superresolutionofunsupervisedctimagesbasedongan AT huiyanzhao 4superresolutionofunsupervisedctimagesbasedongan |