Hyperpolarized 13 C metabolic imaging of the human abdomen with spatiotemporal denoising

PURPOSE

Improving the quality and maintaining the fidelity of large coverage abdominal hyperpolarized (HP) 13 C MRI studies with a patch based global-local higher-order singular value decomposition (GL-HOVSD) spatiotemporal denoising approach.

METHODS

Denoising performance was first evaluated using the simulated [1-13 C]pyruvate dynamics at different noise levels to determine optimal kglobal and klocal parameters. The GL-HOSVD spatiotemporal denoising method with the optimized parameters was then applied to two HP [1-13 C]pyruvate EPI abdominal human cohorts (n = 7 healthy volunteers and n = 8 pancreatic cancer patients).

RESULTS

The parameterization of kglobal  = 0.2 and klocal  = 0.9 denoises abdominal HP data while retaining image fidelity when evaluated by RMSE. The kPX (conversion rate of pyruvate-to-metabolite, X = lactate or alanine) difference was shown to be <20% with respect to ground-truth metabolic conversion rates when there is adequate SNR (SNRAUC  > 5) for downstream metabolites. In both human cohorts, there was a greater than nine-fold gain in peak [1-13 C]pyruvate, [1-13 C]lactate, and [1-13 C]alanine apparent SNRAUC . The improvement in metabolite SNR enabled a more robust quantification of kPL and kPA . After denoising, we observed a 2.1 ± 0.4 and 4.8 ± 2.5-fold increase in the number of voxels reliably fit across abdominal FOVs for kPL and kPA quantification maps.

CONCLUSION

Spatiotemporal denoising greatly improves visualization of low SNR metabolites particularly [1-13 C]alanine and quantification of [1-13 C]pyruvate metabolism in large FOV HP 13 C MRI studies of the human abdomen.

Multi-scale feature aggregation and fusion network with self-supervised multi-level perceptual loss for textures preserving low-dose CT denoising

OBJECTIVE

The textures and detailed structures in computed tomography (CT) images are highly desirable for clinical diagnosis. This study aims to expand the current body of work on textures and details preserving convolutional neural networks for low-dose CT (LDCT) image denoising task.

APPROACH

This study proposed a novel Multi-scale Feature Aggregation and Fusion network (MFAF-net) for LDCT image denoising. Specifically, we proposed a Multi-scale Residual Feature Aggregation Module (MRFAM) to characterize multi-scale structural information in CT images, which captures regional-specific inter-scale variations using learned weights. We further proposed a Cross-level Feature Fusion Module (CFFM) to integrate cross-level features, which adaptively weights the contributions of features from encoder to decoder by using a Spatial Pyramid Attention (SPA) mechanism. Moreover, we proposed a Self-supervised Multi-level Perceptual Loss Module (SMPLM) to generate multi-level auxiliary perceptual supervision for recovery of salient textures and structures of tissues and lesions in CT images, which takes advantage of abundant semantic information at various levels. We introduced parameters for the perceptual loss to adaptively weight the contributions of auxiliary features of different levels and we also introduced an automatic parameter tuning strategy for these parameters.

MAIN RESULTS

Extensive experimental studies were performed to validate the effectiveness of the proposed method. Experimental results demonstrate that the proposed method can achieve better performance on both fine textures preservation and noise suppression for CT image denoising task compared with other competitive CNN based methods.

SIGNIFICANCE

The proposed MFAF-net takes advantage of multi-scale receptive fields, cross-level features integration and self-supervised multi-level perceptual loss, enabling more effective recovering of fine textures and detailed structures of tissues and lesions in CT images.

PETformer network enables ultra-low-dose total-body PET imaging without structural prior

Objective.Positron emission tomography (PET) is essential for non-invasive imaging of metabolic processes in healthcare applications. However, the use of radiolabeled tracers exposes patients to ionizing radiation, raising concerns about carcinogenic potential, and warranting efforts to minimize doses without sacrificing diagnostic quality.Approach.In this work, we present a novel neural network architecture, PETformer, designed for denoising ultra-low-dose PET images without requiring structural priors such as computed tomography (CT) or magnetic resonance imaging. The architecture utilizes a U-net backbone, synergistically combining multi-headed transposed attention blocks with kernel-basis attention and channel attention mechanisms for both short- and long-range dependencies and enhanced feature extraction. PETformer is trained and validated on a dataset of 317 patients imaged on a total-body uEXPLORER PET/CT scanner.Main results.Quantitative evaluations using structural similarity index measure and liver signal-to-noise ratio showed PETformer's significant superiority over other established denoising algorithms across different dose-reduction factors.Significance.Its ability to identify and recover intrinsic anatomical details from background noise with dose reductions as low as 2% and its capacity in maintaining high target-to-background ratios while preserving the integrity of uptake values of small lesions enables PET-only fast and accurate disease diagnosis. Furthermore, PETformer exhibits computational efficiency with only 37 M trainable parameters, making it well-suited for commercial integration.

High Resolution Multi-delay Arterial Spin Labeling with Transformer based Denoising for Pediatric Perfusion MRI

Multi-delay arterial spin labeling (MDASL) can quantitatively measure cerebral blood flow (CBF) and arterial transit time (ATT), which is particularly suitable for pediatric perfusion imaging. Here we present a high resolution (iso-2mm) MDASL protocol and performed test-retest scans on 21 typically developing children aged 8 to 17 years. We further proposed a Transformer-based deep learning (DL) model with k-space weighted image average (KWIA) denoised images as reference for training the model. The performance of the model was evaluated by the SNR of perfusion images, as well as the SNR, bias and repeatability of the fitted CBF and ATT maps. The proposed method was compared to several benchmark methods including KWIA, joint denoising and reconstruction with total generalized variation (TGV) regularization, as well as directly applying a pretrained Transformer model on a larger dataset. The results show that the proposed Transformer model with KWIA reference can effectively denoise multi-delay ASL images, not only improving the SNR for perfusion images of each delay, but also improving the SNR for the fitted CBF and ATT maps. The proposed method also improved test-retest repeatability of whole-brain perfusion measurements. This may facilitate the use of MDASL in neurodevelopmental studies to characterize typical and aberrant brain development.

A denoising method based on deep learning for proton radiograph using energy resolved dose function

Objective.Proton radiograph has been broadly applied in proton radiotherapy which is affected by scattered protons which result in the lower spatial resolution of proton radiographs than that of x-ray images. Traditional image denoising method may lead to the change of water equivalent path length (WEPL) resulting in the lower WEPL measurement accuracy. In this study, we proposed a new denoising method of proton radiographs based on energy resolved dose function curves.Approach.Firstly, the corresponding relationship between the distortion of WEPL characteristic curve, and energy and proportion of scattered protons was established. Then, to improve the accuracy of proton radiographs, deep learning technique was used to remove scattered protons and correct deviated WEPL values. Experiments on a calibration phantom to prove the effectiveness and feasibility of this method were performed. In addition, an anthropomorphic head phantom was selected to demonstrate the clinical relevance of this technology and the denoising effect was analyzed.Main results.The curves of WEPL profiles of proton radiographs became smoother and deviated WEPL values were corrected. For the calibration phantom proton radiograph, the average absolute error of WEPL values decreased from 2.23 to 1.72, the mean percentage difference of all materials of relative stopping power decreased from 1.24 to 0.39, and the average relative WEPL corrected due to the denoising process was 1.06%. In addition, WEPL values correcting were also observed on the proton radiograph for anthropomorphic head phantom due to this denoising process.Significance.The experiments showed that this new method was effective for proton radiograph denoising and had greater advantages than end-to-end image denoising methods, laying the foundation for the implementation of precise proton radiotherapy.

Decomposition iteration strategy for low-dose CT denoising

In the medical field, computed tomography (CT) is a commonly used examination method, but the radiation generated increases the risk of illness in patients. Therefore, low-dose scanning schemes have attracted attention, in which noise reduction is essential. We propose a purposeful and interpretable decomposition iterative network (DISN) for low-dose CT denoising. This method aims to make the network design interpretable and improve the fidelity of details, rather than blindly designing or using deep CNN architecture. The experiment is trained and tested on multiple data sets. The results show that the DISN method can restore the low-dose CT image structure and improve the diagnostic performance when the image details are limited. Compared with other algorithms, DISN has better quantitative and visual performance, and has potential clinical application prospects.

Removal of phase residues in electron holography

Electron holography provides quantitative phase information regarding the electromagnetic fields and the morphology of micro- to nano-scale samples. A phase image reconstructed numerically from an electron hologram sometimes includes phase residues, i.e. origins of unremovable phase discontinuities, which make it much more difficult to quantitatively analyze local phase values. We developed a method to remove the residues in a phase image by a combination of patching local areas of a hologram and denoising based on machine learning. The small patches for a hologram, which were generated using the spatial frequency information of the own fringe patterns, were pasted at each residue point by an algorithm based on sparse modeling. After successive phase reconstruction, the phase components with no dependency on the vicinity were filtered out by Gaussian process regression. We determined that the phase discontinuities that appeared around phase residues were removed and the phase distributions of an atomic resolution phase image of a Pt nanoparticle were sufficiently restored.

Model-based deep CNN-regularized reconstruction for digital breast tomosynthesis with a task-based CNN image assessment approach

Objective. Digital breast tomosynthesis (DBT) is a quasi-three-dimensional breast imaging modality that improves breast cancer screening and diagnosis because it reduces fibroglandular tissue overlap compared with 2D mammography. However, DBT suffers from noise and blur problems that can lower the detectability of subtle signs of cancers such as microcalcifications (MCs). Our goal is to improve the image quality of DBT in terms of image noise and MC conspicuity.Approach. We proposed a model-based deep convolutional neural network (deep CNN or DCNN) regularized reconstruction (MDR) for DBT. It combined a model-based iterative reconstruction (MBIR) method that models the detector blur and correlated noise of the DBT system and the learning-based DCNN denoiser using the regularization-by-denoising framework. To facilitate the task-based image quality assessment, we also proposed two DCNN tools for image evaluation: a noise estimator (CNN-NE) trained to estimate the root-mean-square (RMS) noise of the images, and an MC classifier (CNN-MC) as a DCNN model observer to evaluate the detectability of clustered MCs in human subject DBTs.Main results. We demonstrated the efficacies of CNN-NE and CNN-MC on a set of physical phantom DBTs. The MDR method achieved low RMS noise and the highest detection area under the receiver operating characteristic curve (AUC) rankings evaluated by CNN-NE and CNN-MC among the reconstruction methods studied on an independent test set of human subject DBTs.Significance. The CNN-NE and CNN-MC may serve as a cost-effective surrogate for human observers to provide task-specific metrics for image quality comparisons. The proposed reconstruction method shows the promise of combining physics-based MBIR and learning-based DCNNs for DBT image reconstruction, which may potentially lead to lower dose and higher sensitivity and specificity for MC detection in breast cancer screening and diagnosis.

Low-dose measurement of electric potential distribution in organic light-emitting diode by phase-shifting electron holography with 3D tensor decomposition

To improve the performance of organic light-emitting diodes (OLEDs), it is essential to understand and control the electric potential in the organic semiconductor layers. Electron holography (EH) is a powerful technique for visualizing the potential distribution with a transmission electron microscope. However, it has a serious issue that high-energy electrons may damage the organic layers, meaning that a low-dose EH is required. Here, we used a machine learning technique, three-dimensional (3D) tensor decomposition, to denoise electron interference patterns (holograms) of bilayer OLEDs composed of N,N'-di-[(1-naphthyl)-N,N'-diphenyl]-(1,1'-biphenyl)-4,4'-diamine (α-NPD) and tris-(8-hydroxyquinoline)aluminum (Alq3), acquired under a low-dose rate of 130 e- nm-2 s-1. The effect of denoising on the phase images reconstructed from the holograms was evaluated in terms of both the phase measurement error and the peak signal-to-noise ratio. We achieved a precision equivalent to that of a conventional measurement that had an exposure time 60 times longer. The electric field within the Alq3 layer decreased as the cumulative dose increased, which indicates that the Alq3 layer was degraded by the electron irradiation. On the basis of the degradation of the electric field, we concluded that the tolerance dose without damaging the OLED sample is about 1.7 × 105 e- nm-2, which is about 0.6 times that of the conventional EH. The combination of EH and 3D tensor decomposition denoising is capable of making a time series measurement of an OLED sample without any effect from the electron irradiation.

Speed Up of Volumetric Non-Local Transform-Domain Filter Utilising HPC Architecture

This paper presents a parallel implementation of a non-local transform-domain filter (BM4D). The effectiveness of the parallel implementation is demonstrated by denoising image series from computed tomography (CT) and magnetic resonance imaging (MRI). The basic idea of the filter is based on grouping and filtering similar data within the image. Due to the high level of similarity and data redundancy, the filter can provide even better denoising quality than current extensively used approaches based on deep learning (DL). In BM4D, cubes of voxels named patches are the essential image elements for filtering. Using voxels instead of pixels means that the area for searching similar patches is large. Because of this and the application of multi-dimensional transformations, the computation time of the filter is exceptionally long. The original implementation of BM4D is only single-threaded. We provide a parallel version of the filter that supports multi-core and many-core processors and scales on such versatile hardware resources, typical for high-performance computing clusters, even if they are concurrently used for the task. Our algorithm uses hybrid parallelisation that combines open multi-processing (OpenMP) and message passing interface (MPI) technologies and provides up to 283× speedup, which is a 99.65% reduction in processing time compared to the sequential version of the algorithm. In denoising quality, the method performs considerably better than recent DL methods on the data type that these methods have yet to be trained on.

Enhancing RABASAR for Multi-Temporal SAR Image Despeckling through Directional Filtering and Wavelet Transform

The presence of speckle noise severely hampers the interpretability of synthetic aperture radar (SAR) images. While research on despeckling single-temporal SAR images is well-established, there remains a significant gap in the study of despeckling multi-temporal SAR images. Addressing the limitations in the acquisition of the "superimage" and the generation of ratio images within the RABASAR despeckling framework, this paper proposes an enhanced framework. This enhanced framework proposes a direction-based segmentation approach for multi-temporal SAR non-local means filtering (DSMT-NLM) to obtain the "superimage". The DSMT-NLM incorporates the concept of directional segmentation and extends the application of the non-local means (NLM) algorithm to multi-temporal images. Simultaneously, the enhanced framework employs a weighted averaging method based on wavelet transform (WAMWT) to generate superimposed images, thereby enhancing the generation process of ratio images. Experimental results demonstrate that compared to RABASAR, Frost, and NLM, the proposed method exhibits outstanding performance. It not only effectively removes speckle noise from multi-temporal SAR images and reduces the generation of false details, but also successfully achieves the fusion of multi-temporal information, aligning with experimental expectations.

Deuterium imaging of the Warburg effect at sub-millimolar concentrations by joint processing of the kinetic and spectral dimensions

Deuterium metabolic imaging (DMI) is a promising molecular MRI approach, which follows the administration of deuterated substrates and their metabolization. [6,6'-2 H2 ]-glucose for instance is preferentially converted in tumors to [3,3'-2 H2 ]-lactate as a result of the Warburg effect, providing a distinct resonance whose mapping using time-resolved spectroscopic imaging can diagnose cancer. The MR detection of low-concentration metabolites such as lactate, however, is challenging. It has been recently shown that multi-echo balanced steady-state free precession (ME-bSSFP) increases the signal-to-noise ratio (SNR) of these experiments approximately threefold over regular chemical shift imaging; the present study examines how DMI's sensitivity can be increased further by advanced processing methods. Some of these, such as compressed sensing multiplicative denoising and block-matching/3D filtering, can be applied to any spectroscopic/imaging methods. Sensitivity-enhancing approaches were also specifically tailored to ME-bSSFP DMI, by relying on priors related to the resonances' positions and to features of the metabolic kinetics. Two new methods are thus proposed that use these constraints for enhancing the sensitivity of both the spectral images and the metabolic kinetics. The ability of these methods to improve DMI is evidenced in pancreatic cancer studies carried at 15.2 T, where suitable implementations of the proposals imparted eightfold or more SNR improvement over the original ME-bSSFP data, at no informational cost. Comparisons with other propositions in the literature are briefly discussed.

Domain-Aware Few-Shot Learning for Optical Coherence Tomography Noise Reduction

Speckle noise has long been an extensively studied problem in medical imaging. In recent years, there have been significant advances in leveraging deep learning methods for noise reduction. Nevertheless, adaptation of supervised learning models to unseen domains remains a challenging problem. Specifically, deep neural networks (DNNs) trained for computational imaging tasks are vulnerable to changes in the acquisition system's physical parameters, such as: sampling space, resolution, and contrast. Even within the same acquisition system, performance degrades across datasets of different biological tissues. In this work, we propose a few-shot supervised learning framework for optical coherence tomography (OCT) noise reduction, that offers high-speed training (of the order of seconds) and requires only a single image, or part of an image, and a corresponding speckle-suppressed ground truth, for training. Furthermore, we formulate the domain shift problem for OCT diverse imaging systems and prove that the output resolution of a despeckling trained model is determined by the source domain resolution. We also provide possible remedies. We propose different practical implementations of our approach, verify and compare their applicability, robustness, and computational efficiency. Our results demonstrate the potential to improve sample complexity, generalization, and time efficiency, for coherent and non-coherent noise reduction via supervised learning models, that can also be leveraged for other real-time computer vision applications.

Structure-preserving low-dose computed tomography image denoising using a deep residual adaptive global context attention network

Background

Low-dose computed tomography (LDCT) scans can effectively reduce the radiation damage to patients, but this is highly detrimental to CT image quality. Deep convolutional neural networks (CNNs) have shown their potential in improving LDCT image quality. However, the conventional CNN-based approaches rely fundamentally on the convolution operations, which are ineffective for modeling the correlations among nonlocal similar structures and the regionally distinct statistical properties in CT images. This modeling deficiency hampers the denoising performance for CT images derived in this manner.

Methods

In this paper, we propose an adaptive global context (AGC) modeling scheme to describe the nonlocal correlations and the regionally distinct statistics in CT images with negligible computation load. We further propose an AGC-based long-short residual encoder-decoder (AGC-LSRED) network for efficient LDCT image noise artifact-suppression tasks. Specifically, stacks of residual AGC attention blocks (RAGCBs) with long and short skip connections are constructed in the AGC-LSRED network, which allows valuable structural and positional information to be bypassed through these identity-based skip connections and thus eases the training of the deep denoising network. For training the AGC-LSRED network, we propose a compound loss that combines the L1 loss, adversarial loss, and self-supervised multi-scale perceptual loss.

Results

Quantitative and qualitative experimental studies were performed to verify and validate the effectiveness of the proposed method. The simulation experiments demonstrated the proposed method exhibits the best result in terms of noise suppression [root-mean-square error (RMSE) =9.02; peak signal-to-noise ratio (PSNR) =33.17] and fine structure preservation [structural similarity index (SSIM) =0.925] compared with other competitive CNN-based methods. The experiments on real data illustrated that the proposed method has advantages over other methods in terms of radiologists' subjective assessment scores (averaged scores =4.34).

Conclusions

With the use of the AGC modeling scheme to characterize the structural information in CT images and of residual AGC-attention blocks with long and short skip connections to ease the network training, the proposed AGC-LSRED method achieves satisfactory results in preserving fine anatomical structures and suppressing noise in LDCT images.

Self-supervised structural similarity-based convolutional neural network for cardiac diffusion tensor image denoising

BACKGROUND

Diffusion tensor imaging (DTI) is a promising technique for non-invasively investigating the myocardial fiber structures of human heart. However, low signal-to-noise ratio (SNR) has been a major limit of cardiac DTI to prevent us from detecting myocardium structure accurately. Therefore, it is important to remove the effect of noise on diffusion weighted (DW) images.

PURPOSE

Although the conventional and deep learning-based denoising methods have shown the potential to deal with effectively the noise in DW images, most of them are redundant information dependent or require the noise-free images as golden standard. In addition, the existed DW image denoising methods often suffer from problems of over-smoothing. To address these issues, we propose a self-supervised learning model, structural similarity based convolutional neural network with edge-weighted loss (SSECNN), to remove the noise effectively in cardiac DTI.

METHODS

Considering that the DW images acquired along different diffusion directions have structural similarity, and the noise in these DW images is independent and identically distributed, the structural similarity-based matching algorithm is proposed to search for the most similar DW images. Such similar noisy DW image pairs are then used as the input and target of the denoising network SSECNN, which consists of several convolutional and residual blocks. Through the self-supervised training with these image pairs, the network can restore the clean DW images and retain the correlations between the denoised DW images along different directions. To avoid the over-smoothing problem, we design a novel edge-weighted loss which enables the network to adaptively adjust the loss weights with iterations and therefore to improve the detail preserve ability of the model. To verify the superiority of the proposed method, comparisons with state-of-the-art (SOTA) denoising methods are performed on both synthetic and real acquired DTI datasets.

RESULTS

Experimental results show that SSECNN can effectively reduce the noise in the DW images while preserving detailed texture and edge information and therefore achieve better performance in DTI reconstruction. For synthetic dataset, compared to the SOTA method, the root mean square error (RMSE), peak signal-to-noise ratio (PSNR), and structure similarity index measure (SSIM) between the denoised DW images obtained with SSECNN and noise-free DW images are improved by 6.94%, 1.98%, and 0.76% respectively when the noise level is 10%. As for the acquired cardiac DTI dataset, the SSECNN method could significantly improve SNR and contrast to noise ratio (CNR) of cardiac DW images and achieve more regular helix angle (HA) and transverse angle (TA) maps. The ablation experimental results validate that using the structure similarity-based method to search the similar DW image pairs yield the smallest loss, and with the help of the edge-weighted loss, the denoised DW images and diffusion metric maps can preserve more details.

CONCLUSIONS

The proposed SSECNN method can fully explore the similarity between the DW images along different diffusion directions. Using such similarity and an edge-weighted loss enable us to denoise cardiac DTI effectively in a self-supervised manner. Our method can overcome the redundancy information dependence and over-smoothing problem of the SOTA methods.

Fast Feature Extraction Method for Brillouin Scattering Spectrum of OPGW Optical Cable Based on BOTDR

Brillouin optical time domain reflectometry (BOTDR) detects fiber temperature and strain data and represents one of the most critical ways of identifying abnormal conditions such as ice coverage and lightning strikes on optical fiber composite overhead ground wire (OPGW) cable. Existing BOTDR extracts brillouin frequency shift (BFS) features with cumulative averaging and curve fitting. BFS feature extraction is slow for long-distance measurements, making realizing real-time measurements on fiber optic cables challenging. We propose a fast feature extraction method for block matching and 3D filtering (BM3D) + Sobel brillouin scattering spectroscopy (BGS). BM3D takes the advantage of non-local means (NLM) and wavelet denoising (WD) and utilizes the spatial-domain non-local principle to enhance the denoising in the transform domain. The global filtering capability of BM3D is utilized to filter out the low cumulative average BGS noise and the BFS feature extraction is completed using Sobel edge detection. Simulation verifies the feasibility of the algorithm, and the proposed method is embedded in BOTDR to measure 30 km of actual OPGW line. The experimental results show that under the same conditions, the processing time of this method is reduced by 37 times compared to that with the 50,000-time cumulative averaging + levenberg marquardt (LM) algorithm without severe distortion of the reference resolution. The method improves the sensor demodulation speed by using image processing technology without changing the existing hardware equipment, which is expected to be widely used in the new generation of BOTDR.

Deep learning-based fluorescence image correction for high spatial resolution precise dosimetry

Objective.While radiation-excited fluorescence imaging has great potential to measure absolute 2D dose distributions with high spatial resolution, the fluorescence images are contaminated by noise or artifacts due to Cherenkov light, scattered light or background noise. This study developed a novel deep learning-based model to correct the fluorescence images for accurate dosimetric application.Approach.181 single-aperture static photon beams were delivered to an acrylic tank containing quinine hemisulfate water solution. The emitted radiation-exited optical signals were detected by a complementary metal-oxide semiconductor camera to acquire fluorescence images with 0.3 × 0.3 mm2pixel size. 2D labels of projected dose distributions were obtained by applying forward projection calculation of the 3D dose distributions calculated by a clinical treatment planning system. To calibrate the projected dose distributions for Cherenkov angular dependency, a novel empirical Cherenkov emission calibration method was performed. Total 400-epoch supervised learning was applied to a convolutional neural network (CNN) model to predict the projected dose distributions from fluorescence images, gantry, and collimator angles. Accuracy of the calculated projected dose distributions was evaluated with that of uncorrected or conventional methods by using a few quantitative evaluation metrics.Main results.The projected dose distributions corrected by the empirical Cherenkov emission calibration represented more accurate noise-free images than the uncalibrated distributions. The proposed CNN model provided accurate projected dose distributions. The mean absolute error of the projected dose distributions was improved from 2.02 to 0.766 mm·Gy by the CNN model correction. Moreover, the CNN correction provided higher gamma index passing rates for three different threshold criteria than the conventional methods.Significance.The deep learning-based method improves the accuracy of dose distribution measurements. This technique will also be applied to optical signal denoising or Cherenkov light discrimination in other imaging modalities. This method will provide an accurate dose verification tool with high spatial resolution.

Context-aware dynamic filtering network for confocal laser endomicroscopy image denoising

Objective.As an emerging diagnosis technology for gastrointestinal diseases, confocal laser endomicroscopy (CLE) is limited by the physical structure of the fiber bundle, leading to the inevitable production of various forms of noise during the imaging process. However, existing denoising methods based on hand-crafted features inefficiently deal with realistic noise in CLE images. To alleviate this challenge, we proposed context-aware kernel estimation and multi-scale dynamic fusion modules to remove realistic noise in CLE images, including multiplicative and additive white noise.Approach.Specifically, a realistic noise statistics model with random noise specific to CLE data is constructed and further used to develop a self-supervised denoised model without the participation of clean images. Secondly, context-aware kernel estimation, which improves the representation of features by similar learnable region weights, addresses the problem of the non-uniform distribution of noises in CLE images and proposes a lightweight denoised model (CLENet). Thirdly, we have developed a multi-scale dynamic fusion module that decouples and recalibrates features, providing a precise and contextually enriched representation of features. Finally, we integrated two developed modules into a U-shaped backbone to build an efficient denoising network named U-CLENet.Main Results.Both proposed methods achieve comparable or better performance with low computational complexity on two gastrointestinal disease CLE image datasets using the same training benchmark.Significance.The proposed approaches improve the visual quality of unclear CLE images for various stages of tumor development, helping to reduce the rate of misdiagnosis in clinical decision-making and achieve computer graphics-assisted diagnosis.

Dose reduction and image enhancement in micro-CT using deep learning

BACKGROUND

In preclinical settings, micro-computed tomography (CT) provides a powerful tool to acquire high resolution anatomical images of rodents and offers the advantage to in vivo non-invasively assess disease progression and therapy efficacy. Much higher resolutions are needed to achieve scale-equivalent discriminatory capabilities in rodents as those in humans. High resolution imaging however comes at the expense of increased scan times and higher doses. Specifically, with preclinical longitudinal imaging, there are concerns that dose accumulation may affect experimental outcomes of animal models.

PURPOSE

Dose reduction efforts under the ALARA (as low as reasonably achievable) principles are thus a key point of attention. However, low dose CT acquisitions inherently induce higher noise levels which deteriorate image quality and negatively impact diagnostic performance. Many denoising techniques already exist, and deep learning (DL) has become increasingly popular for image denoising, but research has mostly focused on clinical CT with limited studies conducted on preclinical CT imaging. We investigate the potential of convolutional neural networks (CNN) for restoring high quality micro-CT images from low dose (noisy) images. The novelty of the CNN denoising frameworks presented in this work consists of utilizing image pairs with realistic CT noise present in the input as well as the target image used for the model training; a noisier image acquired with a low dose protocol is matched to a less noisy image acquired with a higher dose scan of the same mouse.

METHODS

Low and high dose ex vivo micro-CT scans of 38 mice were acquired. Two CNN models, based on a 2D and 3D four-layer U-Net, were trained with mean absolute error (30 training, 4 validation and 4 test sets). To assess denoising performance, ex vivo mice and phantom data were used. Both CNN approaches were compared to existing methods, like spatial filtering (Gaussian, Median, Wiener) and iterative total variation image reconstruction algorithm. Image quality metrics were derived from the phantom images. A first observer study (n = 23) was set-up to rank overall quality of differently denoised images. A second observer study (n = 18) estimated the dose reduction factor of the investigated 2D CNN method.

RESULTS

Visual and quantitative results show that both CNN algorithms exhibit superior performance in terms of noise suppression, structural preservation and contrast enhancement over comparator methods. The quality scoring by 23 medical imaging experts also indicates that the investigated 2D CNN approach is consistently evaluated as the best performing denoising method. Results from the second observer study and quantitative measurements suggest that CNN-based denoising could offer a 2-4× dose reduction, with an estimated dose reduction factor of about 3.2 for the considered 2D network.

CONCLUSIONS

Our results demonstrate the potential of DL in micro-CT for higher quality imaging at low dose acquisition settings. In the context of preclinical research, this offers promising future prospects for managing the cumulative severity effects of radiation in longitudinal studies.

An Improved BM3D Algorithm Based on Image Depth Feature Map and Structural Similarity Block-Matching

We propose an improved BM3D algorithm for block-matching based on UNet denoising network feature maps and structural similarity (SSIM). In response to the traditional BM3D algorithm that directly performs block-matching on a noisy image, without considering the deep-level features of the image, we propose a method that performs block-matching on the feature maps of the noisy image. In this method, we perform block-matching on multiple depth feature maps of a noisy image, and then determine the positions of the corresponding similar blocks in the noisy image based on the block-matching results, to obtain the set of similar blocks that take into account the deep-level features of the noisy image. In addition, we improve the similarity measure criterion for block-matching based on the Structural Similarity Index, which takes into account the pixel-by-pixel value differences in the image blocks while fully considering the structure, brightness, and contrast information of the image blocks. To verify the effectiveness of the proposed method, we conduct extensive comparative experiments. The experimental results demonstrate that the proposed method not only effectively enhances the denoising performance of the image, but also preserves the detailed features of the image and improves the visual quality of the denoised image.

A Residual UNet Denoising Network Based on Multi-Scale Feature Extraction and Attention-Guided Filter

In order to obtain high-quality images, it is very important to remove noise effectively and retain image details reasonably. In this paper, we propose a residual UNet denoising network that adds the attention-guided filter and multi-scale feature extraction blocks. We design a multi-scale feature extraction block as the input block to expand the receiving domain and extract more useful features. We also develop the attention-guided filter block to hold the edge information. Further, we use the global residual network strategy to model residual noise instead of directly modeling clean images. Experimental results show our proposed network performs favorably against several state-of-the-art models. Our proposed model can not only suppress the noise more effectively, but also improve the sharpness of the image.

G-RRDB: An Effective THz Image-Denoising Model for Moldy Wheat

In order to solve the problem of large image noise and unremarkable features caused by factors such as fluctuations in the power of a light source during the terahertz image acquisition of wheat, this paper proposes a THz image-denoising model called G-RRDB. Firstly, a module called Ghost-LKA is proposed by combining a large kernel convolutional attention mechanism module with a Ghost convolutional structure, which improves the characteristics of the network to acquire a global sensory field. Secondly, by integrating a spatial attention mechanism with channel attention, an attention module called DAB is proposed to enhance the network's attention to important features. Thirdly, the Ghost-LKA module and DAB module are combined with the baseline model, thus proposing the dense residual denoising network G-RRDB. Compared with traditional denoising networks, both the PSNR and SSIM are improved. The prediction accuracy of G-RRDB is verified through the classification of the VGG16 network, achieving a rate of 92.8%, which represents an improvement of 1.7% and 0.2% compared to the denoised images obtained from the baseline model and the combined baseline model with the DAB module, respectively. The experimental results demonstrate that G-RRDB, a THz image-denoising model based on dense residual structure for moldy wheat, exhibits excellent denoising performance.

[Diffusion tensor field estimation based on 3D U-Net and diffusion tensor imaging model constraint]

OBJECTIVE

To propose a diffusion tensor field estimation network based on 3D U-Net and diffusion tensor imaging (DTI) model constraint (3D DTI-Unet) to accurately estimate DTI quantification parameters from a small number of diffusion-weighted (DW) images with a low signal-to-noise ratio.

METHODS

The input of 3D DTI-Unet was noisy diffusion magnetic resonance imaging (dMRI) data containing one non-DW image and 6 DW images with different diffusion coding directions. The noise-reduced non-DW image and accurate diffusion tensor field were predicted through 3D U-Net. The dMRI data were reconstructed using the DTI model and compared with the true value of dMRI data to optimize the network and ensure the consistency of the dMRI data with the physical model of the diffusion tensor field. We compared 3D DTI-Unet with two DW image denoising algorithms (MP-PCA and GL-HOSVD) to verify the effect of the proposed method.

RESULTS

The proposed method was better than MP-PCA and GL-HOSVD in terms of quantitative results and visual evaluation of DW images, diffusion tensor field and DTI quantification parameters.

CONCLUSION

The proposed method can obtain accurate DTI quantification parameters from one non-DW image and 6 DW images to reduce image acquisition time and improve the reliability of quantitative diagnosis.

Considering Image Information and Self-Similarity: A Compositional Denoising Network

Recently, convolutional neural networks (CNNs) have been widely used in image denoising, and their performance has been enhanced through residual learning. However, previous research mostly focused on optimizing the network architecture of CNNs, ignoring the limitations of the commonly used residual learning. This paper identifies two of its limitations, which are the neglect of image information and the lack of effective consideration of image self-similarity. To solve these limitations, this paper proposes a compositional denoising network (CDN), which contains two sub-paths, the image information path (IIP) and the noise estimation path (NEP), respectively. IIP is trained via an image-to-image method to extract image information. For NEP, it utilizes image self-similarity from the perspective of training. This similarity-based training method constrains NEP to output similar estimated noise distributions for different image patches with a specific kind of noise. Finally, image information and noise distribution information are comprehensively considered for image denoising. Experimental results indicate that CDN outperforms other CNN-based methods in both synthetic and real-world image denoising, achieving state-of-the-art performance.

Comparison of Training Strategies for Autoencoder-Based Monochromatic Image Denoising

Monochromatic images are used mainly in cases where the intensity of the received signal is examined. The identification of the observed objects as well as the estimation of intensity emitted by them depends largely on the precision of light measurement in image pixels. Unfortunately, this type of imaging is often affected by noise, which significantly degrades the quality of the results. In order to reduce it, numerous deterministic algorithms are used, with Non-Local-Means and Block-Matching-3D being the most widespread and treated as the reference point of the current state-of-the-art. Our article focuses on the utilization of machine learning (ML) for the denoising of monochromatic images in multiple data availability scenarios, including those with no access to noise-free data. For this purpose, a simple autoencoder architecture was chosen and checked for various training approaches on two large and widely used image datasets: MNIST and CIFAR-10. The results show that the method of training as well as architecture and the similarity of images within the image dataset significantly affect the ML-based denoising. However, even without access to any clear data, the performance of such algorithms is frequently well above the current state-of-the-art; therefore, they should be considered for monochromatic image denoising.

Dual-domain fusion deep convolutional neural network for low-dose CT denoising

BACKGROUND

In view of the underlying health risks posed by X-ray radiation, the main goal of the present research is to achieve high-quality CT images at the same time as reducing x-ray radiation. In recent years, convolutional neural network (CNN) has shown excellent performance in removing low-dose CT noise. However, previous work mainly focused on deepening and feature extraction work on CNN without considering fusion of features from frequency domain and image domain.

OBJECTIVE

To address this issue, we propose to develop and test a new LDCT image denoising method based on a dual-domain fusion deep convolutional neural network (DFCNN).

METHODS

This method deals with two domains, namely, the DCT domain and the image domain. In the DCT domain, we design a new residual CBAM network to enhance the internal and external relations of different channels while reducing noise to promote richer image structure information. For the image domain, we propose a top-down multi-scale codec network as a denoising network to obtain more acceptable edges and textures while obtaining multi-scale information. Then, the feature images of the two domains are fused by a combination network.

RESULTS

The proposed method was validated on the Mayo dataset and the Piglet dataset. The denoising algorithm is optimal in both subjective and objective evaluation indexes as compared to other state-of-the-art methods reported in previous studies.

CONCLUSIONS

The study results demonstrate that by applying the new fusion model denoising, denoising results in both image domain and DCT domain are better than other models developed using features extracted in the single image domain.

Denoising digital breast tomosynthesis projections using deep learning with synthetic data as training set

Purpose

Image denoising based on deep neural networks (DNN) needs a big dataset containing digital breast tomosynthesis (DBT) projections acquired in different radiation doses to be trained, which is impracticable. Therefore, we propose extensively investigating the use of synthetic data generated by software for training DNNs to denoise DBT real data.

Approach

The approach consists of generating a synthetic dataset representative of the DBT sample space by software, containing noisy and original images. Synthetic data were generated in two different ways: (a) virtual DBT projections generated by OpenVCT and (b) noisy images synthesized from photography regarding noise models used in DBT (e.g., Poisson-Gaussian noise). Then, DNN-based denoising techniques were trained using a synthetic dataset and tested for denoising physical DBT data. Results were evaluated in quantitative (PSNR and SSIM measures) and qualitative (visual analysis) terms. Furthermore, a dimensionality reduction technique (t-SNE) was used for visualization of sample spaces of synthetic and real datasets.

Results

The experiments showed that training DNN models with synthetic data could denoise DBT real data, achieving competitive results to traditional methods in quantitative terms but showing a better balance between noise filtering and detail preservation in a visual analysis. T-SNE enables us to visualize if synthetic and real noises are in the same sample space.

Conclusion

We propose a solution for the lack of suitable training data to train DNN models for denoising DBT projections, showing that we just need the synthesized noise to be in the same sample space as the target image.

A strategy combining denoising and cryo-EM single particle analysis

In cryogenic electron microscopy (cryo-EM) single particle analysis (SPA), high-resolution three-dimensional structures of biological macromolecules are determined by iteratively aligning and averaging a large number of two-dimensional projections of molecules. Since the correlation measures are sensitive to the signal-to-noise ratio, various parameter estimation steps in SPA will be disturbed by the high-intensity noise in cryo-EM. However, denoising algorithms tend to damage high frequencies and suppress mid- and high-frequency contrast of micrographs, which exactly the precise parameter estimation relies on, therefore, limiting their application in SPA. In this study, we suggest combining a cryo-EM image processing pipeline with denoising and maximizing the signal's contribution in various parameter estimation steps. To solve the inherent flaws of denoising algorithms, we design an algorithm named MScale to correct the amplitude distortion caused by denoising and propose a new orientation determination strategy to compensate for the high-frequency loss. In the experiments on several real datasets, the denoised particles are successfully applied in the class assignment estimation and orientation determination tasks, ultimately enhancing the quality of biomacromolecule reconstruction. The case study on classification indicates that our strategy not only improves the resolution of difficult classes (up to 5 Å) but also resolves an additional class. In the case study on orientation determination, our strategy improves the resolution of the final reconstructed density map by 0.34 Å compared with conventional strategy. The code is available at https://github.com/zhanghui186/Mscale.

Fast and Accurate Gamma Imaging System Calibration Based on Deep Denoising Networks and Self-Adaptive Data Clustering

Gamma imagers play a key role in both industrial and medical applications. Modern gamma imagers typically employ iterative reconstruction methods in which the system matrix (SM) is a key component to obtain high-quality images. An accurate SM could be acquired from an experimental calibration step with a point source across the FOV, but at a cost of long calibration time to suppress noise, posing challenges to real-world applications. In this work, we propose a time-efficient SM calibration approach for a 4π-view gamma imager with short-time measured SM and deep-learning-based denoising. The key steps include decomposing the SM into multiple detector response function (DRF) images, categorizing DRFs into multiple groups with a self-adaptive K-means clustering method to address sensitivity discrepancy, and independently training separate denoising deep networks for each DRF group. We investigate two denoising networks and compare them against a conventional Gaussian filtering method. The results demonstrate that the denoised SM with deep networks faithfully yields a comparable imaging performance with the long-time measured SM. The SM calibration time is reduced from 1.4 h to 8 min. We conclude that the proposed SM denoising approach is promising and effective in enhancing the productivity of the 4π-view gamma imager, and it is also generally applicable to other imaging systems that require an experimental calibration step.

Stopping-power ratio estimation for proton radiotherapy using dual-energy computed tomography and prior-image constrained denoising

BACKGROUND

Dual-energy computed tomography (DECT) is a promising technique for estimating stopping-power ratio (SPR) for proton therapy planning. It is known, however, that deriving electron density (ED) and effective atomic number (EAN) from DECT data can cause noise amplification in the resulting SPR images. This can negate the benefits of DECT.

PURPOSE

This work introduces a new algorithm for estimating SPR from DECT with noise suppression, using a pair of CT scans with spectral separation. The method is demonstrated using phantom measurements.

MATERIALS AND METHODS

An iterative algorithm is presented, reconstructing ED and EAN with noise suppression, based on Prior Image Constrained Denoising (PIC-D). The algorithm is tested using a Siemens Definition AS+ CT scanner (Siemens Healthcare, Forchheim, Germany). Three phantoms are investigated: a calibration phantom (CIRS 062M), a QA phantom (CATPHAN 700), and an anthropomorphic head phantom (CIRS 731-HN). A task-transfer function (TTF) and the noise power spectrum are derived from SPR images of the QA phantom for the evaluation of image quality. Comparisons of accuracy and noise for ED, EAN, and SPR are made for various versions of the algorithm in comparison to a solution based on Siemens syngo.via Rho/Z software and the current clinical standard of a single-energy CT stoichiometric calibration. A gamma analysis is also applied to the SPR images of the head phantom and water-equivalent distance (WED) is evaluated in a treatment planning system for a proton treatment field.

RESULTS

The algorithm is effective at suppressing noise in both ED and EAN and hence also SPR. The noise is tunable to a level equivalent to or lower than that of the syngo.via Rho/Z software. The spatial resolution (10% and 50% frequencies in the TTF) does not degrade even for the highest noise suppression investigated, although the average spatial frequency of noise does decrease. The PIC-D algorithm showed better accuracy than syngo.via Rho/Z for low density materials. In the calibration phantom, it was superior even when excluding lung substitutes, with root-mean-square deviations for ED and EAN less than 0.3% and 2%, respectively, compared to 0.5% and 3%. In the head phantom, however, the SPR accuracy of the PIC-D algorithm was comparable (excluding sinus tissue) to that derived from syngo.via Rho/Z: less than 1% error for soft tissue, brain, and trabecular bone substitutes and 5-7% for cortical bone, with the larger error for the latter likely related to the phantom geometry. Gamma evaluation showed that PIC-D can suppress noise in a patient-like geometry without introducing substantial errors in SPR. The absolute pass rates were almost identical for PIC-D and syngo.via Rho/Z. In the WED evaluations no general differences were shown.

CONCLUSIONS

The PIC-D DECT algorithm provides scanner-specific calibration and tunable noise suppression. It is vendor agnostic and applicable to any pair of CT scans with spectral separation. Improved accuracy to current methods was not clearly demonstrated for the complex geometry of a head phantom, but the suppression of noise without spatial resolution degradation and the possibility of incorporating constraints on image properties, suggests the usefulness of the approach.

Vision Transformers in Image Restoration: A Survey

The Vision Transformer (ViT) architecture has been remarkably successful in image restoration. For a while, Convolutional Neural Networks (CNN) predominated in most computer vision tasks. Now, both CNN and ViT are efficient approaches that demonstrate powerful capabilities to restore a better version of an image given in a low-quality format. In this study, the efficiency of ViT in image restoration is studied extensively. The ViT architectures are classified for every task of image restoration. Seven image restoration tasks are considered: Image Super-Resolution, Image Denoising, General Image Enhancement, JPEG Compression Artifact Reduction, Image Deblurring, Removing Adverse Weather Conditions, and Image Dehazing. The outcomes, the advantages, the limitations, and the possible areas for future research are detailed. Overall, it is noted that incorporating ViT in the new architectures for image restoration is becoming a rule. This is due to some advantages compared to CNN, such as better efficiency, especially when more data are fed to the network, robustness in feature extraction, and a better feature learning approach that sees better the variances and characteristics of the input. Nevertheless, some drawbacks exist, such as the need for more data to show the benefits of ViT over CNN, the increased computational cost due to the complexity of the self-attention block, a more challenging training process, and the lack of interpretability. These drawbacks represent the future research direction that should be targeted to increase the efficiency of ViT in the image restoration domain.

Denoising of BOTDR Dynamic Strain Measurement Using Convolutional Neural Networks

The Brillouin optical time domain reflectometry (BOTDR) system measures the distributed strain and temperature information along the optic fibre by detecting the Brillouin gain spectra (BGS) and finding the Brillouin frequency shift profiles. By introducing small gain stimulated Brillouin scattering (SBS), dynamic measurement using BOTDR can be realized, but the performance is limited due to the noise of the detected information. An image denoising method using the convolutional neural network (CNN) is applied to the derived Brillouin gain spectrum images to enhance the performance of the Brillouin frequency shift detection and the strain vibration measurement of the BOTDR system. By reducing the noise of the BGS images along the length of the fibre under test with different network depths and epoch numbers, smaller frequency uncertainties are obtained, and the sine-fitting R-squared values of the detected strain vibration profiles are also higher. The Brillouin frequency uncertainty is improved by 24% and the sine-fitting R-squared value of the obtained strain vibration profile is enhanced to 0.739, with eight layers of total depth and 200 epochs.

Multi-task generative adversarial network for retinal optical coherence tomography image denoising

Objective. Optical coherence tomography (OCT) has become an essential imaging modality for the assessment of ophthalmic diseases. However, speckle noise in OCT images obscures subtle but important morphological details and hampers its clinical applications. In this work, a novel multi-task generative adversarial network (MGAN) is proposed for retinal OCT image denoising.Approach. To strengthen the preservation of retinal structural information in the OCT denoising procedure, the proposed MGAN integrates adversarial learning and multi-task learning. Specifically, the generator of MGAN simultaneously undertakes two tasks, including the denoising task and the segmentation task. The segmentation task aims at the generation of the retinal segmentation map, which can guide the denoising task to focus on the retina-related region based on the retina-attention module. In doing so, the denoising task can enhance the attention to the retinal region and subsequently protect the structural detail based on the supervision of the structural similarity index measure loss.Main results. The proposed MGAN was evaluated and analyzed on three public OCT datasets. The qualitative and quantitative comparisons show that the MGAN method can achieve higher image quality, and is more effective in both speckle noise reduction and structural information preservation than previous denoising methods.Significance. We have presented a MGAN for retinal OCT image denoising. The proposed method provides an effective way to strengthen the preservation of structural information while suppressing speckle noise, and can promote the OCT applications in the clinical observation and diagnosis of retinopathy.

A task-specific deep-learning-based denoising approach for myocardial perfusion SPECT

Deep-learning (DL)-based methods have shown significant promise in denoising myocardial perfusion SPECT images acquired at low dose. For clinical application of these methods, evaluation on clinical tasks is crucial. Typically, these methods are designed to minimize some fidelity-based criterion between the predicted denoised image and some reference normal-dose image. However, while promising, studies have shown that these methods may have limited impact on the performance of clinical tasks in SPECT. To address this issue, we use concepts from the literature on model observers and our understanding of the human visual system to propose a DL-based denoising approach designed to preserve observer-related information for detection tasks. The proposed method was objectively evaluated on the task of detecting perfusion defect in myocardial perfusion SPECT images using a retrospective study with anonymized clinical data. Our results demonstrate that the proposed method yields improved performance on this detection task compared to using low-dose images. The results show that by preserving task-specific information, DL may provide a mechanism to improve observer performance in low-dose myocardial perfusion SPECT.

RDASNet: Image Denoising via a Residual Dense Attention Similarity Network

In recent years, thanks to the performance advantages of convolutional neural networks (CNNs), CNNs have been widely used in image denoising. However, most of the CNN-based image-denoising models cannot make full use of the redundancy of image data, which limits the expressiveness of the model. We propose a new image-denoising model that aims to extract the local features of the image through CNN and focus on the global information of the image through the attention similarity module (ASM), especially the global similarity details of the image. Furthermore, dilation convolution is used to enlarge the receptive field to better focus on the global features. Moreover, avg-pooling is used to smooth and suppress noise in the ASM to further improve model performance. In addition, through global residual learning, the effect is enhanced from shallow to deep layers. A large number of experiments show that our proposed model has a better image-denoising effect, including quantitative and visual results. It is more suitable for complex blind noise and real images.

A densely connected LDCT image denoising network based on dual-edge extraction and multi-scale attention under compound loss

BACKGROUND

Low dose computed tomography (LDCT) uses lower radiation dose, but the reconstructed images contain higher noise that can have negative impact in disease diagnosis. Although deep learning with the edge extraction operators reserves edge information well, only applying the edge extraction operators to input LDCT images does not yield overall satisfactory results.

OBJECTIVE

To improve LDCT images quality, this study proposes and tests a dual edge extraction multi-scale attention mechanism convolution neural network (DEMACNN) based on a compound loss.

METHODS

The network uses edge extraction operators to extract edge information from both the input images and the feature maps in the network, improving the utilization of the edge operators and retaining the images edge information. The feature enhancement block is constructed by fusing the attention mechanism and multi-scale module, enhancing effective information, while suppressing useless information. The residual learning method is used to learn the network, improving the performance of the network, and solving the problem of gradient disappearance. Except for the network structure, a compound loss function, which consists of the MSE loss, the proposed joint total variation loss, and the edge loss, is proposed to enhance the denoising ability of the network and reserve the edge of images.

RESULTS

Compared with other advanced methods (REDCNN, CT-former and EDCNN), the proposed new network achieves the best PSNR and SSIM values in LDCT images of the abdomen, which are 33.3486 and 0.9104, respectively. In addition, the new network also performs well on head and chest image data.

CONCLUSION

The experimental results demonstrate that the proposed new network structure and denoising algorithm not only effectively removes the noise in LDCT images, but also protects the edges and details of the images well.

Serial intravital 2-photon microscopy and analysis of the kidney using upright microscopes

Serial intravital 2-photon microscopy of the kidney and other abdominal organs is a powerful technique to assess tissue function and structure simultaneously and over time. Thus, serial intravital microscopy can capture dynamic tissue changes during health and disease and holds great potential to characterize (patho-) physiological processes with subcellular resolution. However, successful image acquisition and analysis require significant expertise and impose multiple potential challenges. Abdominal organs are rhythmically displaced by breathing movements which hamper high-resolution imaging. Traditionally, kidney intravital imaging is performed on inverted microscopes where breathing movements are partly compensated by the weight of the animal pressing down. Here, we present a custom and easy-to-implement setup for intravital imaging of the kidney and other abdominal organs on upright microscopes. Furthermore, we provide image processing protocols and a new plugin for the free image analysis software FIJI to process multichannel fluorescence microscopy data. The proposed image processing pipelines cover multiple image denoising algorithms, sample drift correction using 2D registration, and alignment of serial imaging data collected over several weeks using landmark-based 3D registration. The provided tools aim to lower the barrier of entry to intravital microscopy of the kidney and are readily applicable by biomedical practitioners.

Truncated total variation in fractional B-spline wavelet transform for micro-CT image denoising

BACKGROUND

In medical applications, computed tomography (CT) is widely used to evaluate various sample characteristics. However, image quality of CT reconstruction can be degraded due to artifacts.

OBJECTIVE

To propose and test a truncated total variation (truncation TV) model to solve the problem of large penalties for the total variation (TV) model.

METHODS

In this study, a truncated TV image denoising model in the fractional B-spline wavelet domain is developed to obtain the best solution. The method is validated by the analysis of CT reconstructed images of actual biological Pigeons samples. For this purpose, several indices including the peak signal-to-noise ratio (PSNR), structural similarity index (SSIM) and mean square error (MSE) are used to evaluate the quality of images.

RESULTS

Comparing to the conventional truncated TV model that yields 22.55, 0.688 and 361.17 in PSNR, SSIM and MSE, respectively, using the proposed fractional B-spline-truncated TV model, the computed values of these evaluation indices change to 24.24, 0.898 and 244.98, respectively, indicating substantial reduction of image noise with higher PSNR and SSIM, and lower MSE.

CONCLUSIONS

Study results demonstrate that compared with many classic image denoising methods, the new denoising algorithm proposed in this study can more effectively suppresses the reconstructed CT image artifacts while maintaining the detailed image structure.

NG-GAN: A Robust Noise-Generation Generative Adversarial Network for Generating Old-Image Noise

Numerous old images and videos were captured and stored under unfavorable conditions. Hence, old images and videos have uncertain and different noise patterns compared with those of modern ones. Denoising old images is an effective technique for reconstructing a clean image containing crucial information. However, obtaining noisy-clean image pairs for denoising old images is difficult and challenging for supervised learning. Preparing such a pair is expensive and burdensome, as existing denoising approaches require a considerable number of noisy-clean image pairs. To address this issue, we propose a robust noise-generation generative adversarial network (NG-GAN) that utilizes unpaired datasets to replicate the noise distribution of degraded old images inspired by the CycleGAN model. In our proposed method, the perception-based image quality evaluator metric is used to control noise generation effectively. An unpaired dataset is generated by selecting clean images with features that match the old images to train the proposed model. Experimental results demonstrate that the dataset generated by our proposed NG-GAN can better train state-of-the-art denoising models by effectively denoising old videos. The denoising models exhibit significantly improved peak signal-to-noise ratios and structural similarity index measures of 0.37 dB and 0.06 on average, respectively, on the dataset generated by our proposed NG-GAN.

Low-dose CT denoising with a high-level feature refinement and dynamic convolution network

BACKGROUND

Since the potential health risks of the radiation generated by computer tomography (CT), concerns have been expressed on reducing the radiation dose. However, low-dose CT (LDCT) images contain complex noise and artifacts, bringing uncertainty to medical diagnosis.

PURPOSE

Existing deep learning (DL)-based denoising methods are difficult to fully exploit hierarchical features of different levels, limiting the effect of denoising. Moreover, the standard convolution kernel is parameter sharing and cannot be adjusted dynamically with input change. This paper proposes an LDCT denoising network using high-level feature refinement and multiscale dynamic convolution to mitigate these problems.

METHODS

The dual network structure proposed in this paper consists of the feature refinement network (FRN) and the dynamic perception network (DPN). The FDN extracts features of different levels through residual dense connections. The high-level hierarchical information is transmitted to DPN to improve the low-level representations. In DPN, the two networks' features are fused by local channel attention (LCA) to assign weights in different regions and handle CT images' delicate tissues better. Then, the dynamic dilated convolution (DDC) with multibranch and multiscale receptive fields is proposed to enhance the expression and processing ability of the denoising network. The experiments were trained and tested on the dataset "NIH-AAPM-Mayo Clinic Low-Dose CT Grand Challenge," consisting of 10 anonymous patients with normal-dose abdominal CT and LDCT at 25% dose. In addition, external validation was performed on the dataset "Low Dose CT Image and Projection Data," which included 300 chest CT images at 10% dose and 300 head CT images at 25% dose.

RESULTS

Proposed method compared with seven mainstream LDCT denoising algorithms. On the Mayo dataset, achieved peak signal-to-noise ratio (PSNR): 46.3526 dB (95% CI: 46.0121-46.6931 dB) and structural similarity (SSIM): 0.9844 (95% CI: 0.9834-0.9854). Compared with LDCT, the average increase was 3.4159 dB and 0.0239, respectively. The results are relatively optimal and statistically significant compared with other methods. In external verification, our algorithm can cope well with ultra-low-dose chest CT images at 10% dose and obtain PSNR: 28.6130 (95% CI: 28.1680-29.0580 dB) and SSIM: 0.7201 (95% CI: 0.7101-0.7301). Compared with LDCT, PSNR/SSIM is increased by 3.6536dB and 0.2132, respectively. In addition, the quality of LDCT can also be improved in head CT denoising.

CONCLUSIONS

This paper proposes a DL-based LDCT denoising algorithm, which utilizes high-level features and multiscale dynamic convolution to optimize the network's denoising effect. This method can realize speedy denoising and performs well in noise suppression and detail preservation, which can be helpful for the diagnosis of LDCT.

Disentangling Noise from Images: A Flow-Based Image Denoising Neural Network

The prevalent convolutional neural network (CNN)-based image denoising methods extract features of images to restore the clean ground truth, achieving high denoising accuracy. However, these methods may ignore the underlying distribution of clean images, inducing distortions or artifacts in denoising results. This paper proposes a new perspective to treat image denoising as a distribution learning and disentangling task. Since the noisy image distribution can be viewed as a joint distribution of clean images and noise, the denoised images can be obtained via manipulating the latent representations to the clean counterpart. This paper also provides a distribution-learning-based denoising framework. Following this framework, we present an invertible denoising network, FDN, without any assumptions on either clean or noise distributions, as well as a distribution disentanglement method. FDN learns the distribution of noisy images, which is different from the previous CNN-based discriminative mapping. Experimental results demonstrate FDN's capacity to remove synthetic additive white Gaussian noise (AWGN) on both category-specific and remote sensing images. Furthermore, the performance of FDN surpasses that of previously published methods in real image denoising with fewer parameters and faster speed.

Segmentation-guided Denoising Network for Low-dose CT Imaging

BACKGROUND

To reduce radiation exposure and improve diagnosis in low-dose computed tomography, several deep learning (DL)-based image denoising methods have been proposed to suppress noise and artifacts over the past few years. However, most of them seek an objective data distribution approximating the gold standard and neglect structural semantic preservation. Moreover, the numerical response in CT images presents substantial regional anatomical differences among tissues in terms of X-ray absorbency.

METHODS

In this paper, we introduce structural semantic information for low-dose CT imaging. First, the regional segmentation prior to low-dose CT can guide the denoising process. Second the structural semantical results can be considered as evaluation metrics on the estimated normal-dose CT images. Then, a semantic feature transform is engaged to combine the semantic and image features on a semantic fusion module. In addition, the structural semantic loss function is introduced to measure the segmentation difference.

RESULTS

Experiments are conducted on clinical abdomen data obtained from a clinical hospital, and the semantic labels consist of subcutaneous fat, muscle and visceral fat associated with body physical evaluation. Compared with other DL-based methods, the proposed method achieves better performance on quantitative metrics and better semantic evaluation.

CONCLUSIONS

The quantitative experimental results demonstrate the promising performance of the proposed methods in noise reduction and structural semantic preservation. While, the proposed method may suffer from several limitations on abnormalities, unknown noise and different manufacturers. In the future, the proposed method will be further explored, and wider applications in PET/CT and PET/MR will be sought.

Non-Local Mean Denoising Algorithm Based on Fractional Compact Finite Difference Scheme Effectively Reduces Speckle Noise in Optical Coherence Tomography Images

Optical coherence tomography (OCT) is used in various fields such, as medical diagnosis and material inspection, as a non-invasive and high-resolution optical imaging modality. However, an OCT image is damaged by speckle noise during its generation, thus reducing the image quality. To address this problem, a non-local means (NLM) algorithm based on the fractional compact finite difference scheme (FCFDS) is proposed to remove the speckle noise in OCT images. FCFDS uses more local pixel information when compared to integer-order difference operators. The FCFDS operator is introduced into the NLM algorithm to construct a high-precision weight calculation so that the proposed algorithm can effectively reduce the speckle noise in the OCT images. Experiments on simulations and real OCT images show that the proposed method is comparable to other state-of-the-art despeckling methods and can substantially reduce noise and preserve image details such as edges and structures. Speckle noise removal can further promote the application of the proposed algorithm in medical diagnosis and industrial detection, as it has key research value.

A Single-Image Noise Estimation Algorithm Based on Pixel-Level Low-Rank Low-Texture Patch and Principal Component Analysis

Noise level is an important parameter for image denoising in many image-processing applications. We propose a noise estimation algorithm based on pixel-level low-rank, low-texture subblocks and principal component analysis for white Gaussian noise. First, an adaptive clustering algorithm, based on a dichotomy merge, adaptive pixel-level low-rank matrix construction method and a gradient covariance low-texture subblock selection method, is proposed to construct a pixel-level low-rank, low-texture subblock matrix. The adaptive clustering algorithm can improve the low-rank property of the constructed matrix and reduce the content of the image information in the eigenvalues of the matrix. Then, an eigenvalue selection method is proposed to eliminate matrix eigenvalues representing the image to avoid an inaccurate estimation of the noise level caused by using the minimum eigenvalue. The experimental results show that, compared with existing state-of-the-art methods, our proposed algorithm has, in most cases, the highest accuracy and robustness of noise level estimation for various scenarios with different noise levels, especially when the noise is high.

Denoise Functional Magnetic Resonance Imaging With Random Matrix Theory Based Principal Component Analysis

High-resolution functional MRI (fMRI) is largely hindered by random thermal noise. Random matrix theory (RMT)-based principal component analysis (PCA) is promising to reduce such noise in fMRI data. However, there is no consensus about the optimal strategy and practice in implementation. In this work, we propose a comprehensive RMT-based denoising method that consists of 1) rank and noise estimation based on a set of newly derived multiple criteria, and 2) optimal singular value shrinkage, with each module explained and implemented based on the RMT. By incorporating the variance stabilizing approach, the denoising method can deal with low signal-to-noise ratio (SNR) (such as <5) magnitude fMRI data with favorable performance compared to other state-of-the-art methods. Results from both simulation and in-vivo high-resolution fMRI data show that the proposed denoising method dramatically improves image restoration quality, promoting functional sensitivity at the same level of functional mapping blurring compared to existing denoising methods. Moreover, the denoising method can serve as a drop-in step in data preprocessing pipelines along with other procedures aimed at removal of structured physiological noises. We expect that the proposed denoising method will play an important role in leveraging high-quality, high-resolution task fMRI, which is desirable in many neuroscience and clinical applications.

Denoising Single Images by Feature Ensemble Revisited

Image denoising is still a challenging issue in many computer vision subdomains. Recent studies have shown that significant improvements are possible in a supervised setting. However, a few challenges, such as spatial fidelity and cartoon-like smoothing, remain unresolved or decisively overlooked. Our study proposes a simple yet efficient architecture for the denoising problem that addresses the aforementioned issues. The proposed architecture revisits the concept of modular concatenation instead of long and deeper cascaded connections, to recover a cleaner approximation of the given image. We find that different modules can capture versatile representations, and a concatenated representation creates a richer subspace for low-level image restoration. The proposed architecture's number of parameters remains smaller than in most of the previous networks and still achieves significant improvements over the current state-of-the-art networks.

Low-Light Image Enhancement Based on Constraint Low-Rank Approximation Retinex Model

Images captured in a low-light environment are strongly influenced by noise and low contrast, which is detrimental to tasks such as image recognition and object detection. Retinex-based approaches have been continuously explored for low-light enhancement. Nevertheless, Retinex decomposition is a highly ill-posed problem. The estimation of the decomposed components should be combined with proper constraints. Meanwhile, the noise mixed in the low-light image causes unpleasant visual effects. To address these problems, we propose a Constraint Low-Rank Approximation Retinex model (CLAR). In this model, two exponential relative total variation constraints were imposed to ensure that the illumination is piece-wise smooth and that the reflectance component is piece-wise continuous. In addition, the low-rank prior was introduced to suppress the noise in the reflectance component. With a tailored separated alternating direction method of multipliers (ADMM) algorithm, the illumination and reflectance components were updated accurately. Experimental results on several public datasets verify the effectiveness of the proposed model subjectively and objectively.

EMDS-6: Environmental Microorganism Image Dataset Sixth Version for Image Denoising, Segmentation, Feature Extraction, Classification, and Detection Method Evaluation

Environmental microorganisms (EMs) are ubiquitous around us and have an important impact on the survival and development of human society. However, the high standards and strict requirements for the preparation of environmental microorganism (EM) data have led to the insufficient of existing related datasets, not to mention the datasets with ground truth (GT) images. This problem seriously affects the progress of related experiments. Therefore, This study develops the Environmental Microorganism Dataset Sixth Version (EMDS-6), which contains 21 types of EMs. Each type of EM contains 40 original and 40 GT images, in total 1680 EM images. In this study, in order to test the effectiveness of EMDS-6. We choose the classic algorithms of image processing methods such as image denoising, image segmentation and object detection. The experimental result shows that EMDS-6 can be used to evaluate the performance of image denoising, image segmentation, image feature extraction, image classification, and object detection methods. EMDS-6 is available at the https://figshare.com/articles/dataset/EMDS6/17125025/1.

Framework for denoising Monte Carlo photon transport simulations using deep learning

SIGNIFICANCE

The Monte Carlo (MC) method is widely used as the gold-standard for modeling light propagation inside turbid media, such as human tissues, but combating its inherent stochastic noise requires one to simulate a large number photons, resulting in high computational burdens.

AIM

We aim to develop an effective image denoising technique using deep learning (DL) to dramatically improve the low-photon MC simulation result quality, equivalently bringing further acceleration to the MC method.

APPROACH

We developed a cascade-network combining DnCNN with UNet, while extending a range of established image denoising neural-network architectures, including DnCNN, UNet, DRUNet, and deep residual-learning for denoising MC renderings (ResMCNet), in handling three-dimensional MC data and compared their performances against model-based denoising algorithms. We also developed a simple yet effective approach to creating synthetic datasets that can be used to train DL-based MC denoisers.

RESULTS

Overall, DL-based image denoising algorithms exhibit significantly higher image quality improvements over traditional model-based denoising algorithms. Among the tested DL denoisers, our cascade network yields a 14 to 19 dB improvement in signal-to-noise ratio, which is equivalent to simulating 25  ×   to 78  ×   more photons. Other DL-based methods yielded similar results, with our method performing noticeably better with low-photon inputs and ResMCNet along with DRUNet performing better with high-photon inputs. Our cascade network achieved the highest quality when denoising complex domains, including brain and mouse atlases.

CONCLUSIONS

Incorporating state-of-the-art DL denoising techniques can equivalently reduce the computation time of MC simulations by one to two orders of magnitude. Our open-source MC denoising codes and data can be freely accessed at http://mcx.space/.

INSIDEnet: Interpretable nonexpansive data-efficient network for denoising in grating interferometry breast CT

PURPOSE

Breast cancer is the most common malignancy in women. Unfortunately, current breast imaging techniques all suffer from certain limitations: they are either not fully three-dimensional, have insufficient resolution or low soft-tissue contrast. Grating Interferometry Breast Computed Tomography (GI-BCT) is a promising X-ray phase contrast modality that could overcome these limitations by offering high soft-tissue contrast and excellent 3D resolution. To enable the transition of this technology to clinical practice, dedicated data processing algorithms must be developed in order to effectively retrieve the signals of interest from the measured raw data.

METHODS

This article proposes a novel denoising algorithm which can cope with the high noise amplitudes and heteroscedasticity which arise in GI-BCT when operated in a low-dose regime to effectively regularize the ill-conditioned GI-BCT inverse problem. We present a data-driven algorithm called INSIDEnet which combines different ideas such as multiscale image processing, transform-domain filtering, transform learning and explicit orthogonality to build an Interpretable NonexpanSIve Data-Efficient network (INSIDEnet).

RESULTS

We apply the method to simulated breast phantom datasets and to real data acquired on a GI-BCT prototype and show that the proposed algorithm outperforms traditional state-of-the-art filters and is competitive with deep neural networks. The strong inductive bias given by the proposed model's architecture allows to reliably train the algorithm with very limited data while providing high model interpretability, thus offering a great advantage over classical convolutional neural networks (CNNs).

CONCLUSIONS

The proposed INSIDEnet is highly data-efficient, interpretable and outperforms state-of-the-art CNNs when trained on very limited training data. We expect the proposed method to become an important tool as part of a dedicated Plug-and-Play GI-BCT reconstruction framework, needed to translate this promising technology to the clinics.