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

Published on Jul 9, 2020in arXiv: Image and Video Processing
Mayank Patwari2
Estimated H-index: 2
(FAU: University of Erlangen-Nuremberg),
Ralf Gutjahr7
Estimated H-index: 7
+ 1 AuthorsAndreas Maier39
Estimated H-index: 39
Sources
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.
📖 Papers frequently viewed together
2019
2 Authors (Sheng He, Genke Yang)
2017
3 Authors (Tianyang Wang, ..., Kaoning Hu)
References10
Newest
#1Mayank Patwari (FAU: University of Erlangen-Nuremberg)H-Index: 2
#2Ralf Gutjahr (Siemens)H-Index: 7
Last. Andreas Maier (FAU: University of Erlangen-Nuremberg)H-Index: 39
view all 4 authors...
With the increasing use of CT in diagnostic imaging, reducing the clinical radiation dose is necessary for ensuring patient safety. Reduced radiation dose results in quantum noise which adversely affects image quality and diagnostic value. Moreover, obtaining high quality images to act as reference images for image quality assessment is difficult. Therefore, automatic no-reference quality assessment of reconstructed images is necessary to preserve diagnostic image quality, while controlling radi...
Source
Low-dose computed tomography (LDCT) has attracted major attention in the medical imaging field, since CT-associated X-ray radiation carries health risks for patients. The reduction of the CT radiation dose, however, compromises the signal-to-noise ratio, which affects image quality and diagnostic performance. Recently, deep-learning-based algorithms have achieved promising results in LDCT denoising, especially convolutional neural network (CNN) and generative adversarial network (GAN) architectu...
Source
#1Chenyang Shen (UTSW: University of Texas Southwestern Medical Center)H-Index: 11
#2Yesenia Gonzalez (UTSW: University of Texas Southwestern Medical Center)H-Index: 6
Last. Xun Jia (UTSW: University of Texas Southwestern Medical Center)H-Index: 45
view all 5 authors...
A number of image-processing problems can be formulated as optimization problems. The objective function typically contains several terms specifically designed for different purposes. Parameters in front of these terms are used to control the relative importance among them. It is of critical importance to adjust these parameters, as quality of the solution depends on their values. Tuning parameters are a relatively straight forward task for a human, as one can intuitively determine the direction...
Source
#1Jelmer M. Wolterink (UU: Utrecht University)H-Index: 24
#2Tim Leiner (UU: Utrecht University)H-Index: 59
Last. Ivana Išgum (UU: Utrecht University)H-Index: 52
view all 4 authors...
Noise is inherent to low-dose CT acquisition. We propose to train a convolutional neural network (CNN) jointly with an adversarial CNN to estimate routine-dose CT images from low-dose CT images and hence reduce noise. A generator CNN was trained to transform low-dose CT images into routine-dose CT images using voxelwise loss minimization. An adversarial discriminator CNN was simultaneously trained to distinguish the output of the generator from routine-dose CT images. The performance of this dis...
Source
#1Franck VidalH-Index: 45
Source
#1Michael ManhartH-Index: 8
#2Rebecca FahrigH-Index: 33
Last. Andreas Maier (FAU: University of Erlangen-Nuremberg)H-Index: 39
view all 5 authors...
● We created a joint bilateral filter for energy-selective detectors with encouraging first results. ● The SNR was improved from 3.3 to 72.3, while a low rRMSE is preserved and only little cross-talk between the channels is introduced. Guided Noise Reduction for Spectral CT with Energy-Selective Photon Counting Detectors Michael Manhart1,2, Rebecca Fahrig3, Joachim Hornegger1,4, Arnd Doerfler2, Andreas Maier1,4 1 Pattern Recognition Lab, Department of Computer Science, Friedrich-Alexander Univer...
#1Xiang Zhu (UCSC: University of California, Santa Cruz)H-Index: 10
#2Peyman Milanfar (UCSC: University of California, Santa Cruz)H-Index: 65
Across the field of inverse problems in image and video processing, nearly all algorithms have various parameters which need to be set in order to yield good results. In practice, usually the choice of such parameters is made empirically with trial and error if no “ground-truth” reference is available. Some analytical methods such as cross-validation and Stein's unbiased risk estimate (SURE) have been successfully used to set such parameters. However, these methods tend to be strongly reliant on...
Source
#1Armando Manduca (Mayo Clinic)H-Index: 59
#2Lifeng Yu (Mayo Clinic)H-Index: 45
Last. Joel G. Fletcher (Mayo Clinic)H-Index: 86
view all 7 authors...
Purpose: To investigate a novel locally adaptive projection space denoising algorithm for low-dose CT data. Methods: The denoising algorithm is based on bilateral filtering, which smooths values using a weighted average in a local neighborhood, with weights determined according to both spatial proximity and intensity similarity between the center pixel and the neighboring pixels. This filtering is locally adaptive and can preserve important edge information in the sinogram, thus maintaining high...
Source
#1Zhou Wang (Center for Neural Science)H-Index: 1
#2Alan C. Bovik (University of Texas at Austin)H-Index: 106
Last. Eero P. SimoncelliH-Index: 86
view all 4 authors...
Objective methods for assessing perceptual image quality traditionally attempted to quantify the visibility of errors (differences) between a distorted image and a reference image using a variety of known properties of the human visual system. Under the assumption that human visual perception is highly adapted for extracting structural information from a scene, we introduce an alternative complementary framework for quality assessment based on the degradation of structural information. As a spec...
Source
#1Peter F.C. Gilbert (LMB: Laboratory of Molecular Biology)H-Index: 13
Abstract A method of reconstruction (ART) has recently been proposed (Gordon, Bender & Herman, 1970) which consists in iteratively changing a trial structure until its projections are consistent with the original projections of the unknown structure. It is shown that in general ART produces erroneous reconstructions. An alternative iterative method is proposed which will give correct reconstructions under certain conditions. One of the potential applications of this method is in determining the ...
Source
Cited By0
Newest
This website uses cookies.
We use cookies to improve your online experience. By continuing to use our website we assume you agree to the placement of these cookies.
To learn more, you can find in our Privacy Policy.