Limited parameter denoising for low-dose X-ray computed tomography using deep reinforcement learning

Abstract

Background: The use of deep learning has successfully solved several problems in the field of medical imaging. Deep learning has been applied to the CT denoising problem successfully. However, the use of deep learning requires large amounts of data to train deep convolutional networks (CNNs). Moreover, due to the large parameter count, such deep CNNs may cause unexpected results.

Purpose: In this study, we introduce a novel CT denoising framework, which has interpretable behavior and provides useful results with limited data.

Methods: We employ bilateral filtering in both the projection and volume domains to remove noise. To account for nonstationary noise, we tune the σ parameters of the volume for every projection view and every volume pixel. The tuning is carried out by two deep CNNs. Due to the impracticality of labeling, the two-deep CNNs are trained via a Deep-Q reinforcement learning task. The reward for the task is generated by using a custom reward function represented by a neural network. Our experiments were carried out on abdominal scans for the Mayo Clinic the cancer imaging archive (TCIA) dataset and the American association of physicists in medicine (AAPM) Low Dose CT Grand Challenge.

Results: Our denoising framework has excellent denoising performance increasing the peak signal to noise ratio (PSNR) from 28.53 to 28.93 and increasing the structural similarity index (SSIM) from 0.8952 to 0.9204. We outperform several state-of-the-art deep CNNs, which have several orders of magnitude higher number of parameters (p-value [PSNR] = 0.000, p-value [SSIM] = 0.000). Our method does not introduce any blurring, which is introduced by mean squared error (MSE) loss-based methods, or any deep learning artifacts, which are introduced by wasserstein generative adversarial network (WGAN)-based models. Our ablation studies show that parameter tuning and using our reward network results in the best possible results.

Conclusions: We present a novel CT denoising framework, which focuses on interpretability to deliver good denoising performance, especially with limited data. Our method outperforms state-of-the-art deep neural networks. Future work will be focused on accelerating our method and generalizing it to different geometries and body parts.

Keywords: computed tomography; image reconstruction; reinforcement learning.