Low-Dose CT Image Reconstruction With a Deep Learning Prior

Hyoung Suk Park,Kiwan Jeon,Kyungsang Kim

doi:10.1109/access.2020.3020040

Abstract

In low-dose computed tomography (LDCT), a penalized weighted least squares (PWLS) approach that incorporates the Poisson statistics of X-ray photons can significantly reduce excessive quantum noise. To improve the quality of LDCT images, prior information such as the total variation, Markov random field, and nonlocal mean, can be imposed onto the target image. However, this information may be limited to reflect the characteristics of the target images, thereby resulting in unexpected side effects (e.g. blurry images). In this paper, we propose a PWLS method combined with a deep learning prior, which is learned from standard dose CT (SDCT) images. The proposed model learns a noise reduction function that maps an LDCT image to its corresponding SDCT images and estimates the prior distribution using a Pearson χ 2 divergence. The model can be converted to the least squares generative adversarial network with an added PWLS objective, where the optimal generator acts as the noise reduction function. We reformulate the proposed model as a constrained optimization problem and solve it using the alternating optimization (AO) algorithm. Clinical SDCT and simulated LDCT scans of ten patients were used to show the validity of the proposed method. Results show that the proposed method outperforms other PWLS methods, by imposing priors such as the total variation and the nonlocal mean.

Highlights

X-ray computed tomography (X-ray CT) is a diagnostic imaging modality widely used in medical radiology
The results showed that the proposed method achieves the best peak signal to noise ratio (PSNR), structure similarity (SSIM), and mean square error (MSE) values, whereas the penalized weighted least squares (PWLS)-nonlocal means (NLM) method shows the worst performance in terms of PSNR
We estimated the prior distribution of the target images in the sense of Pearson χ 2 divergence using many standard dose CT (SDCT) samples

Summary

Introduction

X-ray computed tomography (X-ray CT) is a diagnostic imaging modality widely used in medical radiology. It produces cross-sectional images of the human body with high resolution. In clinical practice, decreasing X-ray current is the most common method of reducing the X-ray dose [15]. The quality of low-dose CT (LDCT) is partly limited by excessive quantum noise, which results in the degradation of diagnostic performance. Over the last few decades, numerous iterative reconstruction (IR) methods for noise reduction have been proposed. Some of these methods have incorporated statistical methods such as Maximum a Posteriori (MAP) approach for data fitting in a sinogram space.

Methods

Results

Conclusion