Simultaneous super resolution and denoising in X-ray CT using deep learning

Abstract

Background

It is common for X-ray computed tomography (CT) images to be reconstructed differently for various clinical examination purposes. This is primarily because we aim to meet two clinical requirements: improving spatial resolution and reducing noise by changing the reconstruction parameters.

Objective

Two deep learning-based methods, super resolution (SR) and denoising, respectively, have been proposed to address these requests. We present a single neural network that can perform SR and denoising simultaneously.

Methods

We propose using existing ultra-high-resolution CT (UHR-CT) data to achieve high spatial resolution and reduces noise. We generated specific input data, which is normal-resolution and high-noise data, simulated from UHR-CT data. Afterwards, we apply the network to NR-CT data, the resulting method, called SR-Denoise deep-learning reconstruction (DLR). In experiments, we measured modulation transfer function as the quantitative study. We also evaluated the performance using both simulated and real clinical data in NR-CT data, with UHR-CT data serving as the ground truth.

Results

SR-Denoise DLR achieved performance on both SR and denoising tasks that was equivalent to training them individually and outperformed the methods currently used in clinical settings.

Conclusions

SR-Denoise DLR utilizes the spatial resolution of UHR-CT and takes advantage of NR-CT to significantly reduce noise.

Keywords

super resolution noise reduction deep learning reconstruction ultra-high-resolution computed tomography chest CT images

Get full access to this article

View all access options for this article.

References

Yanagawa

Hata

Honda

, et al. Subjective and objective comparisons of image quality between ultra-high-resolution CT and conventional area detector CT in phantoms and cadaveric human lungs. Eur Radiol 2018; 28: 5060–5068.

Iwasawa

Sato

Yamaya

, et al. Ultra-high-resolution computed tomography can demonstrate alveolar collapse in novel coronavirus (COVID-19) pneumonia. Jpn J Radiol 2020; 38: 394–398.

Nakamura

Narita

Higaki

, et al. Diagnostic value of deep learning reconstruction for radiation dose reduction at abdominal ultra-high-resolution CT. Eur Radiol 2021; 31: 4700–4709.

Beysang

Villani

Boubaker

, et al. Ultra-high-resolution CT of the temporal bone: comparison between deep learning reconstruction and hybrid and model-based iterative reconstruction. Diagn Interv Imaging 2024; 105: 233–242.

Liu

Shi

, et al. Computed tomography super-resolution using convolutional neural networks. In: 2017 IEEE International Conference on Image Processing (ICIP), 2017.

Umehara

Ota

Ishida

. Application of super-resolution convolutional neural network for enhancing image resolution in chest CT. J Digit Imaging 2018; 31: 441–450.

Lei

Niu

Zhang

, et al. CT Image denoising and deblurring with deep learning: current status and perspectives. IEEE Trans Radiat Plasma Med Sci 2024; 8: 153–172.

You

Zhang

, et al. CT super-resolution GAN constrained by the identical, residual, and cycle learning ensemble (GAN-CIRCLE). IEEE Trans Med Imaging 2020; 39: 98–101.

Wang

Yoon

, et al. Noise controlled CT super-resolution with conditional diffusion model. In: Proc. The 8th International Conference on Image Formation in X-Ray Computed Tomography, 2024, pp.188–203.

10.

Yuan

Gang

Stayman

. Deep Learning CT Image Restoration using System Blur and Noise Models. arXiv:2407.14983 2024.

11.

Chen

Niu

Gao

, et al. LIT-Former: linking in-plane and through-plane transformers for simultaneous CT image denoising and deblurring. IEEE Trans Med Imaging 2024; 43: 1880–1894.

12.

Hata

Yanagawa

Honda

, et al. Effect of matrix size on the image quality of ultra-high-resolution CT of the lung: comparison of 512 x 512, 1024 x 1024, and 2048 x 2048. Acad Radiol 2018; 25: 869–876.

13.

Boedeker

. AiCE deep learning reconstruction: Bringing the power of ultra-high resolution CT to routine imaging. Technical white paper, Canon Medical Systems Corporation. 2018.

14.

Matsuura

Zhou

Akino

, et al. Feature-aware deep-learning reconstruction for context-sensitive x-ray computed tomography. IEEE Trans Radiat Plasma Med Sci 2021; 5: 99–107.

15.

Hernandez

Shin

Abbey

, et al. Validation of synthesized normal-resolution image data generated from high-resolution acquisitions on a commercial CT scanner. Med Phys 2020; 47: 4775–4785.

16.

Fan

Zamyatin

Nakanishi

. Noise simulation for low-dose computed tomography. In: 2012 IEEE Nuclear Science Symposium and Medical Imaging Conference Record (NSS/MIC), 2012, pp.3641–3643.

17.

Zamyatin

Fan

Schultz

, et al. Validation of compound poisson noise model for computed tomography with energy-integrating detector. In: 2012 IEEE Nuclear Science Symposium and Medical Imaging Conference Record (NSS/MIC), 2012, pp.2689–2691.

18.

Milletari

Navab

Ahmadi

S-A

. V-net: fully convolutional neural networks for volumetric medical image segmentation. In: Proc. 2016 Fourth International Conference on 3D Vision (3DV), 2016, pp.565–571.

19.

Lim

Son

Kim

, et al. Enhanced deep residual networks for single image super-resolution. arXiv:1707.02921 2017.

20.

Wang

Bovik

Sheikh

, et al. Image quality assessment: from error visibility to structural similarity. IEEE Trans Image Process 2004; 13: 600–612.

21.

Niu

Wiedeman

, et al. 3D photon counting CT image super-resolution using conditional diffusion model. 17th International Meeting on Fully 3D Image Reconstruction, arXiv:2408.15283 2023.