Neural Adaptive Restoration of Computed Radiography Images (original) (raw)
On Image Restoration Techniques for Medical Imaging
2007
Abstract In this chapter two new approaches are proposed that address two problems that are commonly encountered in the context of image restoration for medical imaging applications. First, we considered the problem of regularized image restoration when no prior information about the original image and the noise is available. A new paradigm, is adopted according to which the required information is extracted from the available data at the previous iteration step, ie, the partially restored image at each step.
A review on Deep Learning approaches for low-dose Computed Tomography restoration
Complex & Intelligent Systems, 2021
Computed Tomography (CT) is a widely use medical image modality in clinical medicine, because it produces excellent visualizations of fine structural details of the human body. In clinical procedures, it is desirable to acquire CT scans by minimizing the X-ray flux to prevent patients from being exposed to high radiation. However, these Low-Dose CT (LDCT) scanning protocols compromise the signal-to-noise ratio of the CT images because of noise and artifacts over the image space. Thus, various restoration methods have been published over the past 3 decades to produce high-quality CT images from these LDCT images. More recently, as opposed to conventional LDCT restoration methods, Deep Learning (DL)-based LDCT restoration approaches have been rather common due to their characteristics of being data-driven, high-performance, and fast execution. Thus, this study aims to elaborate on the role of DL techniques in LDCT restoration and critically review the applications of DL-based approach...
Spatially-Adaptive Reconstruction in Computed Tomography Using Neural Networks
IEEE Transactions on Medical Imaging, 2015
We propose a supervised machine learning approach for boosting existing signal and image recovery methods and demonstrate its efficacy on example of image reconstruction in computed tomography. Our technique is based on a local nonlinear fusion of several image estimates, all obtained by applying a chosen reconstruction algorithm with different values of its control parameters. Usually such output images have different bias/variance trade-off. The fusion of the images is performed by feed-forward neural network trained on a set of known examples. Numerical experiments show an improvement in reconstruction quality relatively to existing direct and iterative reconstruction methods.
Weight assignment for adaptive image restoration by neural networks
IEEE Transactions on Neural Networks, 2000
This paper presents a scheme for adaptively training the weights, in terms of varying the regularization parameter, in a neural network for the restoration of digital images. The flexibility of neural-network-based image restoration algorithms easily allow the variation of restoration parameters such as blur statistics and regularization value spatially and temporally within the image. This paper focuses on spatial variation of the regularization parameter. We first show that the previously proposed neural-network method based on gradient descent can only find suboptimal solutions, and then introduce a regional processing approach based on local statistics. A method is presented to vary the regularization parameter spatially. This method is applied to a number of images degraded by various levels of noise, and the results are examined. The method is also applied to an image degraded by spatially variant blur. In all cases, the proposed method provides visually satisfactory results in an efficient way.
Learned Shrinkage Approach For Low-Dose Reconstruction in Computed Tomography
We propose a direct nonlinear reconstruction algorithm for Computed Tomography (CT), designed to handle low-dose measurements. It involves the Filtered Back-Projection and adaptive non-linear filtering in both the projection and the image domains. The filter is an extension of the learned shrinkage method by Hel-Or and Shaked to the case of indirect observations. The shrinkage functions are learned using a training set of reference CT images. The optimization is performed with respect to an error functional in the image domain, that combines the Mean Square Error with a gradient-based penalty, promoting image sharpness. Our numerical simulations indicate that the proposed algorithm can manage well with noisy measurements, allowing a dose-reduction by a factor of 4, while reducing noise and streak artifacts in the FBP reconstruction, comparable to the performance of a statistically-based iterative algorithm.
Assigning Automatic Regularization Parameters in Image Restoration
Computer Vision Theory and Applications, 2009
This work aims to define and experimentally evaluate an adaptive strategy based on neural learning to select an appropriate regularization parameter within a regularized restoration process. The appropriate setting of the regularization parameter within the restoration process is a difficult task attempting to achieve an optimal balance between removing edge ringing effects and suppressing additive noise. In this context,in an
IEEE Transactions on Biomedical Engineering, 2000
Repeated X-ray computed tomography (CT) scans are often required in several specific applications such as perfusion imaging, image-guided biopsy needle, image-guided intervention, and radiotherapy with noticeable benefits. However, the associated cumulative radiation dose significantly increases as comparison with that used in the conventional CT scan, which has raised major concerns in patients. In this study, to realize radiation dose reduction by reducing the X-ray tube current and exposure time (mAs) in repeated CT scans, we propose a prior-image induced nonlocal (PINL) regularization for statistical iterative reconstruction via the penalized weighted least-squares (PWLS) criteria, which we refer to as "PWLS-PINL". Specifically, the PINL regularization utilizes the redundant information in the prior image and the weighted least-squares term considers a data-dependent variance estimation, aiming to improve current low-dose image quality. Subsequently, a modified iterative successive overrelaxation algorithm is adopted to optimize the associative objective function. Experimental results on both phantom and patient data show that the present PWLS-PINL method can achieve promising gains over the other existing methods in terms of the noise reduction, low-contrast object detection, and edge detail preservation.
IEEE Transactions on Biomedical Engineering, 2000
In this paper, we propose a novel multiscale penalized weighted least-squares (PWLS) method for restoration of lowdose computed tomography (CT) sinogram. The method utilizes wavelet transform for the multiscale or multiresolution analysis on the sinogram. Specifically, the Mallat-Zhong's wavelet transform is applied to decompose the sinogram to different resolution levels. At each decomposed resolution level, a PWLS criterion is applied to restore the noise-contaminated wavelet coefficients, where the penalty is adaptive to each resolution scale and the weight is updated by an exponential relationship between the data variance and mean at each scale and location. The proposed PWLS method is based on the observations that 1) noise in the CT sinogram after logarithm transform and calibration can be modeled as signal-dependent variables and the sample variance depends on the sample mean by an exponential relationship; and 2) noise reduction can be more effective when it is adaptive to different resolution levels. The effectiveness of the proposed multiscale PWLS method is validated by both computer simulations and experimental studies. The gain by multiscale approach over single scale means is quantified by noise-resolution tradeoff measures.
Adaptive image restoration using a local neural approach
2007
This work aims at defining and experimentally evaluating an iterative strategy based on neural learning for blind image restoration in the presence of blur and noise. A salient aspect of our solution is the local estimation of the restored image based on gradient descent strategies able to estimate both the blurring function and the regularized terms adaptively. Instead of explicitly defining the values of local regularization parameters through predefined functions, an adaptive learning approach is proposed. The method was evaluated experimentally using a test pattern generated by a function checkerboard in Matlab. To investigate whether the strategy can be considered an alternative to conventional restoration procedures the results were compared with those obtained by a well known neural restoration approach.
Neural network regularization in the problem of few-view computed tomography
Computer Optics, 2022
The computed tomography allows to reconstruct the inner morphological structure of an object without physical destructing. The accuracy of digital image reconstruction directly depends on the measurement conditions of tomographic projections, in particular, on the number of recorded projections. In medicine, to reduce the dose of the patient load there try to reduce the number of measured projections. However, in a few-view computed tomography, when we have a small number of projections, using standard reconstruction algorithms leads to the reconstructed images degradation. The main feature of our approach for few-view tomography is that algebraic reconstruction is being finalized by a neural network with keeping measured projection data because the additive result is in zero space of the forward projection operator. The final reconstruction presents the sum of the additive calculated with the neural network and the algebraic reconstruction. First is an element of zero space of the forward projection operator. The second is an element of orthogonal addition to the zero space. Last is the result of applying the algebraic reconstruction method to a few-angle sinogram. The dependency model between elements of zero space of forward projection operator and algebraic reconstruction is built with neural networks. It demonstrated that realization of the suggested approach allows achieving better reconstruction accuracy and better computation time than state-of-the-art approaches on test data from the Low Dose CT Challenge dataset without increasing reprojection error.
Image Restoration using a Network of Reduced and Regularized Neural Networks
International Journal of Computer Applications, 2012
The aim of this paper is to implement an optimal neural network model to resolve the problem of colour image restoration which consists of retrieving original image degraded by invariant blur and corrupted by random white additive noise. We propose in this paper an algorithm which implements a general network of reduced neural networks model and adaptive regularization. The developed model is based on the original model of zhou and the modified model of Paik and katsaggelos. The adaptive regularization parameter is used in our case when degradation model contains an additive component (additive noise) in order to obtain a compromise between image sharpness and noise elimination. It is chosen using an iterative algorithm which calculates the best value that maximizes the PSNR of the restored image. Our model presents some improvements in terms of complexity and quality of restored images. It is shown by experiments that restored images obtained by the proposed model are better in terms of both numerical measurement and visual quality.
Restoration of Natural Images using Iterative Global and Local Adaptive Learning Scheme
Image restoration is a process of restoring the image from a damaged condition due to natural noise or by any stack of operations on the image. In this paper, we found a way to reduce the effect of noise on images using the combination of sparse learning approach with the help of neural networks. To make the Proposed system effective, initially, some images were trained, which are low noised and are natural and then by using residual internal and external priors network helps in restoring the damaged image. In this paper, we opted for various noised images such as Gaussian noised images, CCD and CMOS noised images for the restoration process. Purposefully we are unifying Sparse learning approach with neural networks and SVD to obtain a better-restored image from the effect of noises. We have tested our proposed approach on various image datasets and made a clear notation of working very extensively when compared with existing schemes. On an average, SSIM and PSNR metrics obtained are 0.9635 and 43.5dB, respectively.
A Neural-Network-Based Convex Regularizer for Image Reconstruction
arXiv (Cornell University), 2022
The emergence of deep-learning-based methods to solve image-reconstruction problems has enabled a significant increase in quality. Unfortunately, these new methods often lack reliability and explainability, and there is a growing interest to address these shortcomings while retaining the boost in performance. In this work, we tackle this issue by revisiting regularizers that are the sum of convex-ridge functions. The gradient of such regularizers is parameterized by a neural network that has a single hidden layer with increasing and learnable activation functions. This neural network is trained within a few minutes as a multistep Gaussian denoiser. The numerical experiments for denoising, CT, and MRI reconstruction show improvements over methods that offer similar reliability guarantees.
2008 IEEE International Joint Conference on Neural Networks (IEEE World Congress on Computational Intelligence), 2008
The problem of restoring images degraded by linear position invariant distortions and noise is solved by means of a L1-norm regularization, which is equivalent to determining a L1norm solution of an overdetermined system of linear equations, which results from a data-fitting term plus a regularization term that are both in L1 norm. This system is solved by means of a gradient-based neural network with a discontinuous activation function, which is ensured to converge to a L1-norm solution of the corresponding system of linear equations.
Low-dose computed tomography image restoration using previous normal-dose scan
Medical Physics, 2011
In current computed tomography (CT) examinations, the associated x-ray radiation dose is of a significant concern to patients and operators. A simple and cost-effective means to perform the examinations is to lower the milliampere-seconds (mAs) or kVp parameter (or delivering less x-ray energy to the body) as low as reasonably achievable in data acquisition. However, lowering the mAs parameter will unavoidably increase data noise and the noise would propagate into the CT image if no adequate noise control is applied during image reconstruction. Since a normal-dose high diagnostic CT image scanned previously may be available in some clinical applications, such as CT perfusion imaging and CT angiography (CTA), this paper presents an innovative way to utilize the normal-dose scan as a priori information to induce signal restoration of the current low-dose CT image series. Methods: Unlike conventional local operations on neighboring image voxels, nonlocal means (NLM) algorithm utilizes the redundancy of information across the whole image. This paper adapts the NLM to utilize the redundancy of information in the previous normal-dose scan and further exploits ways to optimize the nonlocal weights for low-dose image restoration in the NLM framework. The resulting algorithm is called the previous normal-dose scan induced nonlocal means (ndiNLM). Because of the optimized nature of nonlocal weights calculation, the ndiNLM algorithm does not depend heavily on image registration between the current low-dose and the previous normal-dose CT scans. Furthermore, the smoothing parameter involved in the ndiNLM algorithm can be adaptively estimated based on the image noise relationship between the current low-dose and the previous normal-dose scanning protocols. Results: Qualitative and quantitative evaluations were carried out on a physical phantom as well as clinical abdominal and brain perfusion CT scans in terms of accuracy and resolution properties. The gain by the use of the previous normal-dose scan via the presented ndiNLM algorithm is noticeable as compared to a similar approach without using the previous normal-dose scan. Conclusions: For low-dose CT image restoration, the presented ndiNLM method is robust in preserving the spatial resolution and identifying the low-contrast structure. The authors can draw the conclusion that the presented ndiNLM algorithm may be useful for some clinical applications such as in perfusion imaging, radiotherapy, tumor surveillance, etc. V
Loss Functions for Image Restoration With Neural Networks
IEEE Transactions on Computational Imaging, 2017
Neural networks are becoming central in several areas of computer vision and image processing and different architectures have been proposed to solve specific problems. The impact of the loss layer of neural networks, however, has not received much attention in the context of image processing: the default and virtually only choice is 2. In this paper, we bring attention to alternative choices for image restoration. In particular, we show the importance of perceptually-motivated losses when the resulting image is to be evaluated by a human observer. We compare the performance of several losses, and propose a novel, differentiable error function. We show that the quality of the results improves significantly with better loss functions, even when the network architecture is left unchanged.
Penalized-likelihood sinogram restoration for computed tomography
IEEE Transactions on Medical Imaging, 2000
We formulate computed tomography (CT) sinogram preprocessing as a statistical restoration problem in which the goal is to obtain the best estimate of the line integrals needed for reconstruction from the set of noisy, degraded measurements. CT measurement data are degraded by a number of factors-including beam hardening and off-focal radiation-that produce artifacts in reconstructed images unless properly corrected. Currently, such effects are addressed by a sequence of sinogram-preprocessing steps, including deconvolution corrections for off-focal radiation, that have the potential to amplify noise. Noise itself is generally mitigated through apodization of the reconstruction kernel, which effectively ignores the measurement statistics, although in high-noise situations adaptive filtering methods that loosely model data statistics are sometimes applied. As an alternative, we present a general imaging model relating the degraded measurements to the sinogram of ideal line integrals and propose to estimate these line integrals by iteratively optimizing a statistically based objective function. We consider three different strategies for estimating the set of ideal line integrals, one based on direct estimation of ideal "monochromatic" line integrals that have been corrected for single-material beam hardening, one based on estimation of ideal "polychromatic" line integrals that can be readily mapped to monochromatic line integrals, and one based on estimation of ideal transmitted intensities, from which ideal, monochromatic line integrals can be readily estimated. The first two approaches involve maximization of a penalized Poisson-likelihood objective function while the third involves minimization of a quadratic penalized weighted least squares (PWLS) objective applied in the transmitted intensity domain. We find that at low exposure levels typical of those being considered for screening CT, the Poisson-likelihood based approaches outperform the PWLS objective as well as a standard approach based on adaptive filtering followed by deconvolution. At higher exposure levels, the approaches all perform similarly.
An Iterative Conjugate Gradient Regularization Method for Image Restoration
2009
Image restoration is an ill-posed inverse problem, which has been introduced the regularization method to suppress over-amplification. In this paper, we propose to apply the iterative regularization method to the image restoration problem and present a nested iterative method, called iterative conjugate gradient regularization method. Convergence properties are established in detail. Based on (6), we also simultaneously determine the regularization parameter based on the restored image at each step. Simulation results show that the proposed iterative regularization method is feasible and effective for image restoration.
Simultaneous iterative image restoration and evaluation of the regularization parameter
IEEE Transactions on Signal Processing, 1992
A nonlinear regularized iterative image restoration algorithm is proposed, according to which only the noise variance is assumed to be known in advance. The algorithm results from a set theoretic regularization approach, where a bound of the stabilizing functional, and therefore the regularization parameter, are updated at each iteration step. Sufficient conditions for the convergence of the algorithm are derived