Efficient System for Delimitation of Benign and Malignant Breast Masses

In this study, a high-performing scheme is introduced to delimit benign and malignant masses in breast ultrasound images. The proposal is built upon by the Nonlocal Means filter for image quality improvement, an Intuitionistic Fuzzy C-Means local clustering algorithm for superpixel generation with high adherence to the edges, and the DBSCAN algorithm for the global clustering of those superpixels in order to delimit masses' regions. The empirical study was performed using two datasets, both with benign and malignant breast tumors. The quantitative results with respect to the BUSI dataset were JSC≥0.907, DM≥0.913, HD≥7.025, and MCR≤6.431 for benign masses and JSC≥0.897, DM≥0.900, HD≥8.666, and MCR≤8.016 for malignant ones, while the MID dataset resulted in JSC≥0.890, DM≥0.905, HD≥8.370, and MCR≤7.241 along with JSC≥0.881, DM≥0.898, HD≥8.865, and MCR≤7.808 for benign and malignant masses, respectively. These numerical results revealed that our proposal outperformed all the evaluated comparative state-of-the-art methods in mass delimitation. This is confirmed by the visual results since the segmented regions had a better edge delimitation.

Polarimetric SAR Speckle Filtering Using a Nonlocal Weighted LMMSE Filter

Despeckling is a key preprocessing step for applications using PolSAR data in most cases. In this paper, a technique based on a nonlocal weighted linear minimum mean-squared error (NWLMMSE) filter is proposed for polarimetric synthetic aperture radar (PolSAR) speckle filtering. In the process of filtering a pixel by the LMMSE estimator, the idea of nonlocal means is employed to evaluate the weights of the samples in the estimator, based on the statistical equalities between the neighborhoods of the sample pixels and the processed pixel. The NWLMMSE estimator is then derived. In the preliminary processing, an effective step is taken to preclassify the pixels, aiming at preserving point targets and considering the similarity of the scattering mechanisms between pixels in the subsequent filter. A simulated image and two real-world PolSAR images are used for illustration, and the experiments show that this filter is effective in speckle reduction, while effectively preserving strong point targets, edges, and the polarimetric scattering mechanism.

CT Image Reconstruction using NLMfuzzyCD Regularization Method

Aims and scope: Computed Tomography (CT) is one of the most efficient clinical diagnostic tools. The main goal of CT is to reproduce an acceptable reconstructed image of an object (either anatomical or functional behaviour) with the help of a limited set of its projections at different angles.

BACKGROUND

To achieve this goal, one of the most commonly iterative reconstruction algorithm called Maximum Likelihood Expectation Maximization (MLEM) is used.

OBJECTIVE

Although the conventional Maximum Likelihood (ML) algorithm can achieve quality images in CT. However, it still suffers from the optimal smoothing as the number of iterations increase.

METHODS

For solving this problem, in this paper present a novel statistical image reconstruction algorithm for CT, which utilizes a nonlocal means fuzzy complex diffusion as a regularization term for noise reduction and edge preservation.

RESULTS

The proposed model was evaluated on four test cases phantoms.

CONCLUSION

Qualitative and quantitative analyses indicate that the proposed technique has higher efficiency for computed tomography. The proposed method yields significant improvements when compare with the state-of-the-art techniques.

Free-breathing liver fat and R 2 ∗ quantification using motion-corrected averaging based on a nonlocal means algorithm

PURPOSE

To propose a motion-robust chemical shift-encoded (CSE) method with high signal-to-noise (SNR) for accurate quantification of liver proton density fat fraction (PDFF) and R 2 ∗ .

METHODS

A free-breathing multi-repetition 2D CSE acquisition with motion-corrected averaging using nonlocal means (NLM) was proposed. PDFF and R 2 ∗ quantified with 2D CSE-NLM were compared to two alternative 2D techniques: direct averaging and single acquisition (2D 1ave) in a digital phantom. Further, 2D NLM was compared in patients to 3D techniques (standard breath-hold, free-breathing and navigated), and the alternative 2D techniques. A reader study and quantitative analysis (Bland-Altman, correlation analysis, paired Student's t-test) were performed to evaluate the image quality and assess PDFF and R 2 ∗ measurements in regions of interest.

RESULTS

In simulations, 2D NLM resulted in lower standard deviations (STDs) of PDFF (2.7%) and R 2 ∗ (8.2  s - 1 ) compared to direct averaging (PDFF: 3.1%, R 2 ∗ : 13.6  s - 1 ) and 2D 1ave (PDFF: 8.7%, R 2 ∗ : 33.2  s - 1 ). In patients, 2D NLM resulted in fewer motion artifacts than 3D free-breathing and 3D navigated, less signal loss than 2D direct averaging, and higher SNR than 2D 1ave. Quantitatively, the STDs of PDFF and R 2 ∗ of 2D NLM were comparable to those of 2D direct averaging (p>0.05). 2D NLM reduced bias, particularly in R 2 ∗ (-5.73 to -0.36  s - 1 ) that arises in direct averaging (-3.96 to 11.22  s - 1 ) in the presence of motion.

CONCLUSIONS

2D CSE-NLM enables accurate mapping of PDFF and R 2 ∗ in the liver during free-breathing.

Multitemporal SAR Image Despeckling Based on a Scattering Covariance Matrix of Image Patch

This paper presents a despeckling method for multitemporal images acquired by synthetic aperture radar (SAR) sensors. The proposed method uses a scattering covariance matrix of each image patch as the basic processing unit, which can exploit both the amplitude information of each pixel and the phase difference between any two pixels in a patch. The proposed filtering framework consists of four main steps: (1) a prefiltering result of each image is obtained by a nonlocal weighted average using only the information of the corresponding time phase; (2) an adaptively temporal linear filter is employed to further suppress the speckle; (3) the final output of each patch is obtained by a guided filter using both the original speckled data and the filtering result of step 3; and (4) an aggregation step is used to tackle the multiple estimations problem for each pixel. The despeckling experiments conducted on both simulated and real multitemporal SAR datasets reveal the pleasing performance of the proposed method in both suppressing speckle and retaining details, when compared with both advanced single-temporal and multitemporal SAR despeckling techniques.

Laplacian Eigenmaps Network-Based Nonlocal Means Method for MR Image Denoising

Magnetic resonance (MR) images are often corrupted by Rician noise which degrades the accuracy of image-based diagnosis tasks. The nonlocal means (NLM) method is a representative filter in denoising MR images due to its competitive denoising performance. However, the existing NLM methods usually exploit the gray-level information or hand-crafted features to evaluate the similarity between image patches, which is disadvantageous for preserving the image details while smoothing out noise. In this paper, an improved nonlocal means method is proposed for removing Rician noise in MR images by using the refined similarity measures. The proposed method firstly extracts the intrinsic features from the pre-denoised image using a shallow convolutional neural network named Laplacian eigenmaps network (LEPNet). Then, the extracted features are used for computing the similarity in the NLM method to produce the denoised image. Finally, the method noise of the denoised image is utilized to further improve the denoising performance. Specifically, the LEPNet model is composed of two cascaded convolutional layers and a nonlinear output layer, in which the Laplacian eigenmaps are employed to learn the filter bank in the convolutional layers and the Leaky Rectified Linear Unit activation function is used in the final output layer to output the nonlinear features. Due to the advantage of LEPNet in recovering the geometric structure of the manifold in the low-dimension space, the features extracted by this network can facilitate characterizing the self-similarity better than the existing NLM methods. Experiments have been performed on the BrainWeb phantom and the real images. Experimental results demonstrate that among several compared denoising methods, the proposed method can provide more effective noise removal and better details preservation in terms of human vision and such objective indexes as peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM).

Assessment of prior image induced nonlocal means regularization for low-dose CT reconstruction: Change in anatomy

PURPOSE

Repeated computed tomography (CT) scans are prescribed for some clinical applications such as lung nodule surveillance. Several studies have demonstrated that incorporating a high-quality prior image into the reconstruction of subsequent low-dose CT (LDCT) acquisitions can either improve image quality or reduce data fidelity requirements. Our proposed previous normal-dose image induced nonlocal means (ndiNLM) regularization method for LDCT is an example of such a method. However, one major concern with prior image based methods is that they might produce false information when the prior image and the current LDCT image show different structures (for example, if a lung nodule emerges, grows, shrinks, or disappears over time). This study aims to assess the performance of the ndiNLM regularization method in situations with change in anatomy.

METHOD

We incorporated the ndiNLM regularization into the statistical image reconstruction (SIR) framework for reconstruction of subsequent LDCT images. Because of its patch-based search mechanism, a rough registration between the prior image and the current LDCT image is adequate for the SIR-ndiNLM method. We assessed the performance of the SIR-ndiNLM method in lung nodule surveillance for two different scenarios: (a) the nodule was not found in a baseline exam but appears in a follow-up LDCT scan; (b) the nodule was present in a baseline exam but disappears in a follow-up LDCT scan. We further investigated the effect of nodule size on the performance of the SIR-ndiNLM method.

RESULTS

We found that a relatively large search-window (e.g., 33 × 33) should be used for the SIR-ndiNLM method to account for misalignment between the prior image and the current LDCT image, and to ensure that enough similar patches can be found in the prior image. With proper selection of other parameters, experimental results with two patient datasets demonstrated that the SIR-ndiNLM method did not miss true nodules nor introduce false nodules in the lung nodule surveillance scenarios described above. We also found that the SIR-ndiNLM reconstruction shows improved image quality when the prior image is similar to the current LDCT image in anatomy. These gains in image quality might appear small upon visual inspection, but they can be detected using quantitative measures. Finally, the SIR-ndiNLM method also performed well in ultra-low-dose conditions and with different nodule sizes.

CONCLUSIONS

This study assessed the performance of the SIR-ndiNLM method in situations in which the prior image and the current LDCT image show substantial anatomical differences, specifically, changes in lung nodules. The experimental results demonstrate that the SIR-ndiNLM method does not introduce false lung nodules nor miss true nodules, which relieves the concern that this method might produce false information. However, there is insufficient evidence that these findings will hold true for all kinds of anatomical changes.

Non-local diffusion-weighted image super-resolution using collaborative joint information

Due to the clinical durable scanning time and other physical constraints, the spatial resolution of diffusion-weighted magnetic resonance imaging (DWI) is highly limited. Using a post-processing method to improve the resolution of DWI holds the potential to improve the investigation of smaller white-matter structures and to reduce partial volume effects. In the present study, a novel non-local mean super-resolution method was proposed to increase the spatial resolution of DWI datasets. Based on a non-local strategy, joint information from the adjacent scanning directions was taken advantage of through the implementation of a novel weighting scheme. Besides this, an efficient rotationally invariant similarity measure was introduced for further improvement of high-resolution image reconstruction and computational efficiency. Quantitative and qualitative comparisons in synthetic and real DWI datasets demonstrated that the proposed method significantly enhanced the resolution of DWI, and is thus beneficial in improving the estimation accuracy for diffusion tensor imaging as well as high-angular resolution diffusion imaging.

l0 regularization based on a prior image incorporated non-local means for limited-angle X-ray CT reconstruction

PURPOSES

Restricted by the scanning environment in some CT imaging modalities, the acquired projection data are usually incomplete, which may lead to a limited-angle reconstruction problem. Thus, image quality usually suffers from the slope artifacts. The objective of this study is to first investigate the distorted domains of the reconstructed images which encounter the slope artifacts and then present a new iterative reconstruction method to address the limited-angle X-ray CT reconstruction problem.

METHODS

The presented framework of new method exploits the structural similarity between the prior image and the reconstructed image aiming to compensate the distorted edges. Specifically, the new method utilizes l0 regularization and wavelet tight framelets to suppress the slope artifacts and pursue the sparsity. New method includes following 4 steps to (1) address the data fidelity using SART; (2) compensate for the slope artifacts due to the missed projection data using the prior image and modified nonlocal means (PNLM); (3) utilize l0 regularization to suppress the slope artifacts and pursue the sparsity of wavelet coefficients of the transformed image by using iterative hard thresholding (l0W); and (4) apply an inverse wavelet transform to reconstruct image. In summary, this method is referred to as "l0W-PNLM".

RESULTS

Numerical implementations showed that the presented l0W-PNLM was superior to suppress the slope artifacts while preserving the edges of some features as compared to the commercial and other popular investigative algorithms. When the image to be reconstructed is inconsistent with the prior image, the new method can avoid or minimize the distorted edges in the reconstructed images. Quantitative assessments also showed that applying the new method obtained the highest image quality comparing to the existing algorithms.

CONCLUSIONS

This study demonstrated that the presented l0W-PNLM yielded higher image quality due to a number of unique characteristics, which include that (1) it utilizes the structural similarity between the reconstructed image and prior image to modify the distorted edges by slope artifacts; (2) it adopts wavelet tight frames to obtain the first and high derivative in several directions and levels; and (3) it takes advantage of l0 regularization to promote the sparsity of wavelet coefficients, which is effective for the inhibition of the slope artifacts. Therefore, the new method can address the limited-angle CT reconstruction problem effectively and have practical significance.

Applications of nonlocal means algorithm in low-dose X-ray CT image processing and reconstruction: A review

Low-dose X-ray computed tomography (LDCT) imaging is highly recommended for use in the clinic because of growing concerns over excessive radiation exposure. However, the CT images reconstructed by the conventional filtered back-projection (FBP) method from low-dose acquisitions may be severely degraded with noise and streak artifacts due to excessive X-ray quantum noise, or with view-aliasing artifacts due to insufficient angular sampling. In 2005, the nonlocal means (NLM) algorithm was introduced as a non-iterative edge-preserving filter to denoise natural images corrupted by additive Gaussian noise, and showed superior performance. It has since been adapted and applied to many other image types and various inverse problems. This paper specifically reviews the applications of the NLM algorithm in LDCT image processing and reconstruction, and explicitly demonstrates its improving effects on the reconstructed CT image quality from low-dose acquisitions. The effectiveness of these applications on LDCT and their relative performance are described in detail.

Electrocardiogram Signal Denoising Using Extreme-Point Symmetric Mode Decomposition and Nonlocal Means

Electrocardiogram (ECG) signals contain a great deal of essential information which can be utilized by physicians for the diagnosis of heart diseases. Unfortunately, ECG signals are inevitably corrupted by noise which will severely affect the accuracy of cardiovascular disease diagnosis. Existing ECG signal denoising methods based on wavelet shrinkage, empirical mode decomposition and nonlocal means (NLM) cannot provide sufficient noise reduction or well-detailed preservation, especially with high noise corruption. To address this problem, we have proposed a hybrid ECG signal denoising scheme by combining extreme-point symmetric mode decomposition (ESMD) with NLM. In the proposed method, the noisy ECG signals will first be decomposed into several intrinsic mode functions (IMFs) and adaptive global mean using ESMD. Then, the first several IMFs will be filtered by the NLM method according to the frequency of IMFs while the QRS complex detected from these IMFs as the dominant feature of the ECG signal and the remaining IMFs will be left unprocessed. The denoised IMFs and unprocessed IMFs are combined to produce the final denoised ECG signals. Experiments on both simulated ECG signals and real ECG signals from the MIT-BIH database demonstrate that the proposed method can suppress noise in ECG signals effectively while preserving the details very well, and it outperforms several state-of-the-art ECG signal denoising methods in terms of signal-to-noise ratio (SNR), root mean squared error (RMSE), percent root mean square difference (PRD) and mean opinion score (MOS) error index.

Sensitivity Analysis of the Scattering-Based SARBM3D Despeckling Algorithm

Synthetic Aperture Radar (SAR) imagery greatly suffers from multiplicative speckle noise, typical of coherent image acquisition sensors, such as SAR systems. Therefore, a proper and accurate despeckling preprocessing step is almost mandatory to aid the interpretation and processing of SAR data by human users and computer algorithms, respectively. Very recently, a scattering-oriented version of the popular SAR Block-Matching 3D (SARBM3D) despeckling filter, named Scattering-Based (SB)-SARBM3D, was proposed. The new filter is based on the a priori knowledge of the local topography of the scene. In this paper, an experimental sensitivity analysis of the above-mentioned despeckling algorithm is carried out, and the main results are shown and discussed. In particular, the role of both electromagnetic and geometrical parameters of the surface and the impact of its scattering behavior are investigated. Furthermore, a comprehensive sensitivity analysis of the SB-SARBM3D filter against the Digital Elevation Model (DEM) resolution and the SAR image-DEM coregistration step is also provided. The sensitivity analysis shows a significant robustness of the algorithm against most of the surface parameters, while the DEM resolution plays a key role in the despeckling process. Furthermore, the SB-SARBM3D algorithm outperforms the original SARBM3D in the presence of the most realistic scattering behaviors of the surface. An actual scenario is also presented to assess the DEM role in real-life conditions.

PET reconstruction via nonlocal means induced prior

BACKGROUND

The traditional Bayesian priors for maximum a posteriori (MAP) reconstruction methods usually incorporate local neighborhood interactions that penalize large deviations in parameter estimates for adjacent pixels; therefore, only local pixel differences are utilized. This limits their abilities of penalizing the image roughness.

OBJECTIVE

To achieve high-quality PET image reconstruction, this study investigates a MAP reconstruction strategy by incorporating a nonlocal means induced (NLMi) prior (NLMi-MAP) which enables utilizing global similarity information of image.

METHODS

The present NLMi prior approximates the derivative of Gibbs energy function by an NLM filtering process. Specially, the NLMi prior is obtained by subtracting the current image estimation from its NLM filtered version and feeding the residual error back to the reconstruction filter to yield the new image estimation.

RESULTS

We tested the present NLMi-MAP method with simulated and real PET datasets. Comparison studies with conventional filtered backprojection (FBP) and a few iterative reconstruction methods clearly demonstrate that the present NLMi-MAP method performs better in lowering noise, preserving image edge and in higher signal to noise ratio (SNR).

CONCLUSIONS

Extensive experimental results show that the NLMi-MAP method outperforms the existing methods in terms of cross profile, noise reduction, SNR, root mean square error (RMSE) and correlation coefficient (CORR).

Uncertainty estimation in diffusion MRI using the nonlocal bootstrap

In this paper, we propose a new bootstrap scheme, called the nonlocal bootstrap (NLB) for uncertainty estimation. In contrast to the residual bootstrap, which relies on a data model, or the repetition bootstrap, which requires repeated signal measurements, NLB is not restricted by the data structure imposed by a data model and obviates the need for time-consuming multiple acquisitions. NLB hinges on the observation that local imaging information recurs in an image. This self-similarity implies that imaging information coming from spatially distant (nonlocal) regions can be exploited for more effective estimation of statistics of interest. Evaluations using in silico data indicate that NLB produces distribution estimates that are in closer agreement with those generated using Monte Carlo simulations, compared with the conventional residual bootstrap. Evaluations using in vivo data demonstrate that NLB produces results that are in agreement with our knowledge on white matter architecture.

Nonlocal regularization of inverse problems: a unified variational framework

We introduce a unifying energy minimization framework for nonlocal regularization of inverse problems. In contrast to the weighted sum of square differences between image pixels used by current schemes, the proposed functional is an unweighted sum of inter-patch distances. We use robust distance metrics that promote the averaging of similar patches, while discouraging the averaging of dissimilar patches. We show that the first iteration of a majorize-minimize algorithm to minimize the proposed cost function is similar to current nonlocal methods. The reformulation thus provides a theoretical justification for the heuristic approach of iterating nonlocal schemes, which re-estimate the weights from the current image estimate. Thanks to the reformulation, we now understand that the widely reported alias amplification associated with iterative nonlocal methods are caused by the convergence to local minimum of the nonconvex penalty. We introduce an efficient continuation strategy to overcome this problem. The similarity of the proposed criterion to widely used nonquadratic penalties (e.g., total variation and lp semi-norms) opens the door to the adaptation of fast algorithms developed in the context of compressive sensing; we introduce several novel algorithms to solve the proposed nonlocal optimization problem. Thanks to the unifying framework, these fast algorithms are readily applicable for a large class of distance metrics.

Nonlocal regularization for active appearance model: Application to medial temporal lobe segmentation

The human medial temporal lobe (MTL) is an important part of the limbic system, and its substructures play key roles in learning, memory, and neurodegeneration. The MTL includes the hippocampus (HC), amygdala (AG), parahippocampal cortex (PHC), entorhinal cortex, and perirhinal cortex--structures that are complex in shape and have low between-structure intensity contrast, making them difficult to segment manually in magnetic resonance images. This article presents a new segmentation method that combines active appearance modeling and patch-based local refinement to automatically segment specific substructures of the MTL including HC, AG, PHC, and entorhinal/perirhinal cortex from MRI data. Appearance modeling, relying on eigen-decomposition to analyze statistical variations in image intensity and shape information in study population, is used to capture global shape characteristics of each structure of interest with a generative model. Patch-based local refinement, using nonlocal means to compare the image local intensity properties, is applied to locally refine the segmentation results along the structure borders to improve structure delimitation. In this manner, nonlocal regularization and global shape constraints could allow more accurate segmentations of structures. Validation experiments against manually defined labels demonstrate that this new segmentation method is computationally efficient, robust, and accurate. In a leave-one-out validation on 54 normal young adults, the method yielded a mean Dice κ of 0.87 for the HC, 0.81 for the AG, 0.73 for the anterior parts of the parahippocampal gyrus (entorhinal and perirhinal cortex), and 0.73 for the posterior parahippocampal gyrus.

A CANDLE for a deeper in vivo insight

A new Collaborative Approach for eNhanced Denoising under Low-light Excitation (CANDLE) is introduced for the processing of 3D laser scanning multiphoton microscopy images. CANDLE is designed to be robust for low signal-to-noise ratio (SNR) conditions typically encountered when imaging deep in scattering biological specimens. Based on an optimized non-local means filter involving the comparison of filtered patches, CANDLE locally adapts the amount of smoothing in order to deal with the noise inhomogeneity inherent to laser scanning fluorescence microscopy images. An extensive validation on synthetic data, images acquired on microspheres and in vivo images is presented. These experiments show that the CANDLE filter obtained competitive results compared to a state-of-the-art method and a locally adaptive optimized non-local means filter, especially under low SNR conditions (PSNR<8dB). Finally, the deeper imaging capabilities enabled by the proposed filter are demonstrated on deep tissue in vivo images of neurons and fine axonal processes in the Xenopus tadpole brain.

Efficient and robust nonlocal means denoising of MR data based on salient features matching

The nonlocal means (NLM) filter has become a popular approach for denoising medical images due to its excellent performance. However, its heavy computational load has been an important shortcoming preventing its use. NLM works by averaging pixels in nonlocal vicinities, weighting them depending on their similarity with the pixel of interest. This similarity is assessed based on the squared difference between corresponding pixels inside local patches centered at the locations compared. Our proposal is to reduce the computational load of this comparison by checking only a subset of salient features associated to the pixels, which suffice to estimate the actual difference as computed in the original NLM approach. The speedup achieved with respect to the original implementation is over one order of magnitude, and, when compared to more recent NLM improvements for MRI denoising, our method is nearly twice as fast. At the same time, we evidence from both synthetic and in vivo experiments that computing of appropriate salient features make the estimation of NLM weights more robust to noise. Consequently, we are able to improve the outcomes achieved with recent state of the art techniques for a wide range of realistic Signal-to-Noise ratio scenarios like diffusion MRI. Finally, the statistical characterization of the features computed allows to get rid of some of the heuristics commonly used for parameter tuning.