Tensor decomposition and non-local means based spectral CT image denoising

Tensor Methods in Biomedical Image Analysis

In the past decade, tensors have become increasingly attractive in different aspects of signal and image processing areas. The main reason is the inefficiency of matrices in representing and analyzing multimodal and multidimensional datasets. Matrices cannot preserve the multidimensional correlation of elements in higher-order datasets and this highly reduces the effectiveness of matrix-based approaches in analyzing multidimensional datasets. Besides this, tensor-based approaches have demonstrated promising performances. These together, encouraged researchers to move from matrices to tensors. Among different signal and image processing applications, analyzing biomedical signals and images is of particular importance. This is due to the need for extracting accurate information from biomedical datasets which directly affects patient's health. In addition, in many cases, several datasets have been recorded simultaneously from a patient. A common example is recording electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) of a patient with schizophrenia. In such a situation, tensors seem to be among the most effective methods for the simultaneous exploitation of two (or more) datasets. Therefore, several tensor-based methods have been developed for analyzing biomedical datasets. Considering this reality, in this paper, we aim to have a comprehensive review on tensor-based methods in biomedical image analysis. The presented study and classification between different methods and applications can show the importance of tensors in biomedical image enhancement and open new ways for future studies.

Multi-energy CT material decomposition using graph model improved CNN

In spectral CT imaging, the coefficient image of the basis material obtained by the material decomposition technique can estimate the tissue composition, and its accuracy directly affects the disease diagnosis. Although the precision of material decomposition is increased by employing convolutional neural networks (CNN), extracting the non-local features from the CT image is restricted using the traditional CNN convolution operator. A graph model built by multi-scale non-local self-similar patterns is introduced into multi-material decomposition (MMD). We proposed a novel MMD method based on graph edge-conditioned convolution U-net (GECCU-net) to enhance material image quality. The GECCU-net focuses on developing a multi-scale encoder. At the network coding stage, three paths are applied to capture comprehensive image features. The local and non-local feature aggregation (LNFA) blocks are designed to integrate the local and non-local features from different paths. The graph edge-conditioned convolution based on non-Euclidean space excavates the non-local features. A hybrid loss function is defined to accommodate multi-scale input images and avoid over-smoothing of results. The proposed network is compared quantitatively with base CNN models on the simulated and real datasets. The material images generated by GECCU-net have less noise and artifacts while retaining more information on tissue. The Structural SIMilarity (SSIM) of the obtained abdomen and chest water maps reaches 0.9976 and 0.9990, respectively, and the RMSE reduces to 0.1218 and 0.4903 g/cm3. The proposed method can improve MMD performance and has potential applications.

CT image denoising methods for image quality improvement and radiation dose reduction

With the ever-increasing use of computed tomography (CT), concerns about its radiation dose have become a significant public issue. To address the need for radiation dose reduction, CT denoising methods have been widely investigated and applied in low-dose CT images. Numerous noise reduction algorithms have emerged, such as iterative reconstruction and most recently, deep learning (DL)-based approaches. Given the rapid advancements in Artificial Intelligence techniques, we recognize the need for a comprehensive review that emphasizes the most recently developed methods. Hence, we have performed a thorough analysis of existing literature to provide such a review. Beyond directly comparing the performance, we focus on pivotal aspects, including model training, validation, testing, generalizability, vulnerability, and evaluation methods. This review is expected to raise awareness of the various facets involved in CT image denoising and the specific challenges in developing DL-based models.

Impact of Noise Level on the Accuracy of Automated Measurement of CT Number Linearity on ACR CT and Computational Phantoms

Background

Methods for segmentation, i.e., Full-segmentation (FS) and Segmentation-rotation (SR), are proposed for maintaining Computed Tomography (CT) number linearity. However, their effectiveness has not yet been tested against noise.

Objective

This study aimed to evaluate the influence of noise on the accuracy of CT number linearity of the FS and SR methods on American College of Radiology (ACR) CT and computational phantoms.

Material and Methods

This experimental study utilized two phantoms, ACR CT and computational phantoms. An ACR CT phantom was scanned by a 128-slice CT scanner with various tube currents from 80 to 200 mA to acquire various noises, with other constant parameters. The computational phantom was added by different Gaussian noises between 20 and 120 Hounsfield Units (HU). The CT number linearity was measured by the FS and SR methods, and the accuracy of CT number linearity was computed on two phantoms.

Results

The two methods successfully segmented both phantoms at low noise, i.e., less than 60 HU. However, segmentation and measurement of CT number linearity are not accurate on a computational phantom using the FS method for more than 60-HU noise. The SR method is still accurate up to 120 HU of noise.

Conclusion

The SR method outperformed the FS method to measure the CT number linearity due to its endurance in extreme noise.

Deep learning based spectral CT imaging

Spectral computed tomography (CT) has attracted much attention in radiation dose reduction, metal artifacts removal, tissue quantification and material discrimination. The x-ray energy spectrum is divided into several bins, each energy-bin-specific projection has a low signal-noise-ratio (SNR) than the current-integrating counterpart, which makes image reconstruction a unique challenge. Traditional wisdom is to use prior knowledge based iterative methods. However, this kind of methods demands a great computational cost. Inspired by deep learning, here we first develop a deep learning based reconstruction method; i.e., U-net with L_p^p-norm, Total variation, Residual learning, and Anisotropic adaption (ULTRA). Specifically, we emphasize the various multi-scale feature fusion and multichannel filtering enhancement with a denser connection encoding architecture for residual learning and feature fusion. To address the image deblurring problem associated with the L₂²- loss, we propose a general L_p^p-loss, p>0. Furthermore, the images from different energy bins share similar structures of the same object, the regularization characterizing correlations of different energy bins is incorporated into the L_p^p- loss function, which helps unify the deep learning based methods with traditional compressed sensing based methods. Finally, the anisotropically weighted total variation is employed to characterize the sparsity in the spatial-spectral domain to regularize the proposed network In particular, we validate our ULTRA networks on three large-scale spectral CT datasets, and obtain excellent results relative to the competing algorithms. In conclusion, our quantitative and qualitative results in numerical simulation and preclinical experiments demonstrate that our proposed approach is accurate, efficient and robust for high-quality spectral CT image reconstruction.

InNetGAN: Inception Network-Based Generative Adversarial Network for Denoising Low-Dose Computed Tomography

Low-dose Computed Tomography (LDCT) has gained a great deal of attention in clinical procedures due to its ability to reduce the patient's risk of exposure to the X-ray radiation. However, reducing the X-ray dose increases the quantum noise and artifacts in the acquired LDCT images. As a result, it produces visually low-quality LDCT images that adversely affect the disease diagnosing and treatment planning in clinical procedures. Deep Learning (DL) has recently become the cutting-edge technology of LDCT denoising due to its high performance and data-driven execution compared to conventional denoising approaches. Although the DL-based models perform fairly well in LDCT noise reduction, some noise components are still retained in denoised LDCT images. One reason for this noise retention is the direct transmission of feature maps through the skip connections of contraction and extraction path-based DL modes. Therefore, in this study, we propose a Generative Adversarial Network with Inception network modules (InNetGAN) as a solution for filtering the noise transmission through skip connections and preserving the texture and fine structure of LDCT images. The proposed Generator is modeled based on the U-net architecture. The skip connections in the U-net architecture are modified with three different inception network modules to filter out the noise in the feature maps passing over them. The quantitative and qualitative experimental results have shown the performance of the InNetGAN model in reducing noise and preserving the subtle structures and texture details in LDCT images compared to the other state-of-the-art denoising algorithms.

A review on Deep Learning approaches for low-dose Computed Tomography restoration

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 approaches for LDCT restoration. To achieve this aim, different aspects of DL-based LDCT restoration applications were analyzed. These include DL architectures, performance gains, functional requirements, and the diversity of objective functions. The outcome of the study highlights the existing limitations and future directions for DL-based LDCT restoration. To the best of our knowledge, there have been no previous reviews, which specifically address this topic.

A metal artifact reduction scheme in CT by a Poisson fusion sinogram based postprocessing method

OBJECTIVE

To reduce secondary artifactes generated by the current interpolation-based metal artifact reduction (MAR) methods, this study proposes and tests a new Poisson fusion sinogram based metal artifact reduction (FS-MAR) method.

METHODS

The proposed FS-MAR method consists of (1) generating the prior image, (2) forward projecting this prior image and applying the Poisson blending technique to seamlessly replace the metal-affected sinogram of the original projection in the metal projection region (MPR) by the prior image projection to get the corrected metal-free sinogram, and (3) performing the filtered back projection (FBP) on the corrected sinogram and filling the metal image back to the metal-free corrected image to get the final artifact reduced image. Simulated images are calculated by taking clinical metal-free CT images as phantoms and inserting metals during the simulated projection process to get the corresponding metal-affected images by the FBP. After the simulated images are processed by the proposed MAR method, two metrics structural similarity index (SSIM) and peak signal-to-noise ratio (PSNR) are used to evaluate image quality. Finally, visual evaluation is also performed using several real clinical metal-affected images obtained from the Revision Radiology group.

RESULTS

In two testing samples, using FS-MAR method yields the highest SSIM and PSNR of 0.8912 and 30.6693, respectively. Visual evaluation results on both simulated and clinical images also show that using FS-MAR method generates less image artifacts than using the interpolation-based algorithm.

CONCLUSIONS

This study demonstrated that with the same prior image, applying the proposed Poisson FS-MAR method can achieve the higher image quality than using the interpolation-based algorithm.

Dual residual convolutional neural network (DRCNN) for low-dose CT imaging

The excessive radiation doses in the application of computed tomography (CT) technology pose a threat to the health of patients. However, applying a low radiation dose in CT can result in severe artifacts and noise in the captured images, thus affecting the diagnosis. Therefore, in this study, we investigate a dual residual convolution neural network (DRCNN) for low-dose CT (LDCT) imaging, whereby the CT images are reconstructed directly from the sinogram by integrating analytical domain transformations, thus reducing the loss of projection information. With this new framework, feature extraction is performed simultaneously on both the sinogram-domain sub-net and the image-domain sub-net, which utilize the residual shortcut networks and play a complementary role in suppressing the projection noise and reducing image error. This new DRCNN approach helps not only decrease the sinogram noise but also preserve significant structural information. The experimental results of simulated and real projection data demonstrate that our DRCNN achieve superior performance over other state-of-art methods in terms of visual inspection and quantitative metrics. For example, comparing with RED-CNN and DP-ResNet, the value of PSNR using our DRCNN is improved by nearly 3 dB and 1 dB, respectively.

Material decomposition for simulated dual-energy breast computed tomography via hybrid optimization method

BACKGROUND

Dual-energy breast CT reconstruction has a potential application that includes separation of microcalcification from healthy breast tissue for assisting early breast cancer detection.

OBJECTIVE

To investigate and validate the noise suppression algorithm applied in the decomposition of the simulated breast phantom into microcalcification and healthy breast.

METHODS

The proposed hybrid optimization method (HOM) uses a simultaneous algebraic reconstruction technique (SART) output as a prior image, which is then incorporated into the self-adaptive dictionary learning. This self-adaptive dictionary learning seeks each group of patches to faithfully represent the learned dictionary, and the sparsity and non-local similarity of group patches are used to enforce the image regularization term of the prior image. We simulate a numerical phantom by adding different levels of Gaussian noise to test performance of the proposed method.

RESULTS

The mean value of peak signal-to-noise ratio (PSNR), structural similarity (SSIM), and root mean square error (RMSE) for the proposed method are (49.043±1.571), (0.997±0.002), (0.003±0.001) and (51.329±1.998), (0.998±0.002), (0.003±0.001) for 35 kVp and 49 kVp, respectively. The PSNR of the proposed method shows greater improvement over TWIST (5.2%), SART (34.6%), FBP (40.4%) and TWIST (3.7%), SART (39.9%), FBP (50.3%) for 35 kVp and 49 kVp energy images, respectively. For the proposed method, the signal-to-noise ratio (SNR) of decomposed normal breast tissue (NBT) is (22.036±1.535), which exceeded that of TWIST, SART, and FBP by 7.5%, 49.6%, and 96.4%, respectively. The results reveal that the proposed algorithm achieves the best performance in both reconstructed and decomposed images under different levels of noise and the performance is due to the high sparsity and good denoising ability of minimization exploited to solve the convex optimization problem.

CONCLUSIONS

This study demonstrates the potential of applying dual-energy reconstruction in breast CT to detect and separate clustered MCs from healthy breast tissues without noise amplification. Compared to other competing methods, the proposed algorithm achieves the best noise suppression performance for both reconstructed and decomposed images.