Low Dose CT Denoising via Joint Bilateral Filtering and Intelligent Parameter Optimization

TL;DR

This paper introduces a CT image denoising method combining joint bilateral filtering guided by a deep CNN and optimized via deep reinforcement learning, achieving high performance with interpretability and structural preservation.

Contribution

It proposes a novel CT denoising approach that integrates joint bilateral filtering with deep reinforcement learning for parameter tuning, reducing complexity and improving interpretability.

Findings

Outperforms state-of-the-art deep neural networks in denoising quality.

Uses only two parameters, enhancing interpretability.

Effectively preserves structural information in CT images.

Abstract

Denoising of clinical CT images is an active area for deep learning research. Current clinically approved methods use iterative reconstruction methods to reduce the noise in CT images. Iterative reconstruction techniques require multiple forward and backward projections, which are time-consuming and computationally expensive. Recently, deep learning methods have been successfully used to denoise CT images. However, conventional deep learning methods suffer from the 'black box' problem. They have low accountability, which is necessary for use in clinical imaging situations. In this paper, we use a Joint Bilateral Filter (JBF) to denoise our CT images. The guidance image of the JBF is estimated using a deep residual convolutional neural network (CNN). The range smoothing and spatial smoothing parameters of the JBF are tuned by a deep reinforcement learning task. Our actor first chooses a parameter, and subsequently chooses an action to tune the value of the parameter. A reward network is designed to direct the reinforcement learning task. Our denoising method demonstrates good denoising performance, while retaining structural information. Our method significantly outperforms state of the art deep neural networks. Moreover, our method has only two parameters, which makes it significantly more interpretable and reduces the 'black box' problem. We experimentally measure the impact of our intelligent parameter optimization and our reward network. Our studies show that our current setup yields the best results in terms of structural preservation.