Chest X-ray images super-resolution reconstruction via recursive neural network
- PDF / 2,879,284 Bytes
- 15 Pages / 439.37 x 666.142 pts Page_size
- 117 Downloads / 174 Views
Chest X-ray images super-resolution reconstruction via recursive neural network Chao-Yue Zhao 1 & Rui-Sheng Jia 1,2 & Qing-Ming Liu 1 & Xiao-Ying Liu 1 & Hong-Mei Sun 1,2 & Xing-Li Zhang 1,2 Received: 29 February 2020 / Revised: 27 July 2020 / Accepted: 28 August 2020 # Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract
To address the problems of insufficient detail extraction and long training time in the super-resolution reconstruction of chest X-ray images, a method of chest Xray images super-resolution reconstruction using recursive neural network is proposed in this paper. Firstly, this paper designs a lightweight recursive network as the main branch, which solves the problem of training difficulty and timeconsuming. Then, to overcome the lack of detail extraction in chest X-ray image, a detail complementary model is designed as another branch of the network to solve the problem of shallow information loss. Finally, the optimized activation function is used to reduce the loss of texture details and make the reconstructed image more complete and richer. When the scale factor is 2, the experimental results show that compared with other methods based on deep learning, such as the deep recursive neural network (DRCN), the details of chest X-ray images reconstructed by our method are more abundant. Specifically, the average value of PSNR and SSIM were improved by 0.17 dB and 0.0013 respectively. Moreover, the reconstruction speed of the images was increased by about 16% compared with DRCN. Keywords Detail complementary module . Deep learning . Medical images . Recursive neural network . Super-resolution reconstruction
* Rui-Sheng Jia [email protected] * Hong-Mei Sun [email protected]
1
College of Computer Science and Engineering, Shandong University of Science and Technology, Qingdao 266590, China
2
Shandong Province Key Laboratory of Wisdom Mine Information Technology, Shandong University of Science and Technology, Qingdao 266590, China
Multimedia Tools and Applications
1 Introduction Super-resolution reconstruction technology is a hot spot in the field of medical image processing research [16]. The utility of medical images is dependent on its quality, and the high-quality medical images can provide more accurate lesion information. However, this presents a challenge that is not only limited by the physical capabilities, such as X-ray machines, MRI scanners, and other equipment, but also needs to consider the harm of radiation dose to the human body. The high-resolution medical images require large radiative dosages for extended exposure times potentially putting the patient at risk. For example, in X-ray imaging, the main factor affecting image quality and patient radiation risk is radiation dose. Using high radiation dose will cause harm to human body, and then using low radiation dose is easy to cause poor image quality. So, using the super-resolution reconstruction method to reconstruct the medical images can solve this problem, which can save costs, reduce harm to patients and h
Data Loading...
