Segmentation of the heart muscle in 3-D pediatric echocardiographic images

This study aimed to show segmentation of the heart muscle in pediatric echocardiographic images as a preprocessing step for tissue analysis. Transthoracic image sequences (2-D and 3-D volume data, both derived in radiofrequency format, directly after beam forming) were registered in real time from four healthy children over three heart cycles. Three preprocessing methods, based on adaptive filtering, were used to reduce the speckle noise for optimizing the distinction between blood and myocardium, while preserving the sharpness of edges between anatomical structures. The filtering kernel size was linked to the local speckle size and the speckle noise characteristics were considered to define the optimal filter in one of the methods. The filtered 2-D images were thresholded automatically as a first step of segmentation of the endocardial wall. The final segmentation step was achieved by applying a deformable contour algorithm. This segmentation of each 2-D image of the 3-D+time (i.e., 4-D) datasets was related to that of the neighboring images in both time and space. By thus incorporating spatial and temporal information of 3-D ultrasound image sequences, an automated method using image statistics was developed to perform 3-D segmentation of the heart muscle.     

Breast ultrasound image segmentation: a survey

Purpose

Breast cancer is the most common form of cancer among women worldwide. Ultrasound imaging is one of the most frequently used diagnostic tools to detect and classify abnormalities of the breast. Recently, computer-aided diagnosis (CAD) systems using ultrasound images have been developed to help radiologists to increase diagnosis accuracy. However, accurate ultrasound image segmentation remains a challenging problem due to various ultrasound artifacts. In this paper, we investigate approaches developed for breast ultrasound (BUS) image segmentation.

Methods

In this paper, we reviewed the literature on the segmentation of BUS images according to the techniques adopted, especially over the past 10 years. By dividing into seven classes (i.e., thresholding-based, clustering-based, watershed-based, graph-based, active contour model, Markov random field and neural network), we have introduced corresponding techniques and representative papers accordingly.

Results

We have summarized and compared many techniques on BUS image segmentation and found that all these techniques have their own pros and cons. However, BUS image segmentation is still an open and challenging problem due to various ultrasound artifacts introduced in the process of imaging, including high speckle noise, low contrast, blurry boundaries, low signal-to-noise ratio and intensity inhomogeneity

Conclusions

To the best of our knowledge, this is the first comprehensive review of the approaches developed for segmentation of BUS images. With most techniques involved, this paper will be useful and helpful for researchers working on segmentation of ultrasound images, and for BUS CAD system developers.

     

The feasibility of hybrid automatic segmentation of axillary lymph nodes from a 3-D sonogram

The use of manual segmentation of lymph nodes, within an ultrasound image, is challenging due to operator dependency and speckle. A group of 23 healthy female volunteers consented to a short imaging session to capture a maximum of three axillary lymph nodes. A feasibility study was completed using both automatic and manual segmentation techniques to analyze a sample of 45, three-dimensional (3-D) nodal volume sets. Level-set segmentation based on geodesic active contours and shape-space learning based on a level-set segmentation approach was used to capture global node shapes. Most of the image feature based segmentation methods failed; however, a more precise automatic segmentation algorithm was obtained using a superimposed shape model. Shape model based segmentation significantly improved the segmentation compared with standard level sets. The best segmentation results were achieved when an experienced sonographer assisted with setting seed surfaces. The initialization of seed surfaces improved the capture of the global shape and lymphatic vessels.     

Fast and adaptive region merging based on perceptual hashing via multi-thresholding for SAR image segmentation

Due to the serious speckle noise in synthetic aperture radar (SAR) image, segmentation of SAR images is still a challenging problem. In this paper, a novel region merging method based on perceptual hashing is proposed for SAR image segmentation. In the proposed method, perceptual hash algorithm (PHA) is utilized to calculate the degree of similarity between different regions during region merging in SAR image segmentation. After reducing the speckle noise by Lee filter which maintains the sharpness of SAR image, a set of different homogeneous regions is constructed based on multi-thresholding and treated as the input data of region merging. The new contribution of this paper is the combination of multi-thresholding for initial segmentation and perceptual hash method for the adaptive process of region merging, which preserves the texture feature of input images and reduces the time complexity of the proposed method. The experimental results on synthetic and real SAR images show that the proposed algorithm is faster and attains higher-quality segmentation results than the three recent state-of-the-art image segmentation methods.     

Global guidance network for breast lesion segmentation in ultrasound images

Automatic breast lesion segmentation in ultrasound helps to diagnose breast cancer, which is one of the dreadful diseases that affect women globally. Segmenting breast regions accurately from ultrasound image is a challenging task due to the inherent speckle artifacts, blurry breast lesion boundaries, and inhomogeneous intensity distributions inside the breast lesion regions. Recently, convolutional neural networks (CNNs) have demonstrated remarkable results in medical image segmentation tasks. However, the convolutional operations in a CNN often focus on local regions, which suffer from limited capabilities in capturing long-range dependencies of the input ultrasound image, resulting in degraded breast lesion segmentation accuracy. In this paper, we develop a deep convolutional neural network equipped with a global guidance block (GGB) and breast lesion boundary detection (BD) modules for boosting the breast ultrasound lesion segmentation. The GGB utilizes the multi-layer integrated feature map as a guidance information to learn the long-range non-local dependencies from both spatial and channel domains. The BD modules learn additional breast lesion boundary map to enhance the boundary quality of a segmentation result refinement. Experimental results on a public dataset and a collected dataset show that our network outperforms other medical image segmentation methods and the recent semantic segmentation methods on breast ultrasound lesion segmentation. Moreover, we also show the application of our network on the ultrasound prostate segmentation, in which our method better identifies prostate regions than state-of-the-art networks.     

Multiple Resolution Bayesian Segmentation of Ultrasound Images

We propose a novel method for obtaining the maximum a posteriori (MAP) probabilistic segmentation of speckle-laden ultrasound images. Our technique is multiple-resolution based, and relies on the conversion of speckle images with Rayleigh statistics to subsampled images with Gaussian statistics. This conversion reduces computation time, as well as allowing accurate parameter estimation for a probabilistic segmentation algorithm. Results appear to provide improvements over previous techniques in terms of low-contrast detail and accuracy.     

Random Forest-Based Bone Segmentation in Ultrasound

Ultrasound (US) imaging is a safe alternative to radiography for guidance during minimally invasive orthopedic procedures. However, ultrasound is challenging to interpret because of the relatively low signal-to-noise ratio and its inherent speckle pattern that decreases image quality. Here we describe a method for automatic bone segmentation in 2-D ultrasound images using a patch-based random forest classifier and several ultrasound specific features, such as shadowing. We illustrate that existing shadow features are not robust to changes in US acquisition parameters, and propose a novel robust shadow feature. We evaluate the method on several US data sets and report that it favorably compares with existing techniques. We achieve a recall of 0.86 at a precision of 0.82 on a test set of 143 spinal US images.     

Segmentation of the liver in ultrasound: a dynamic texture approach

Objective

The segmentation of ultrasound (US) images is useful for several applications in computer aided interventions including the registration of pre-operative CT or MRI to intra-operative US. Shadowing, intensity inhomogeneity and speckle are the common effects on US images. They render the segmentation algorithms developed for other modalities inappropriate due to poor robustness. We present a novel method for classification of hepatic structures including vasculature and liver parenchyma on US images.

Methods

The method considers B-mode US images as a dynamic texture. The dynamics of each pixel are modelled as an auto regressive (AR) process perturbed with Gaussian noise. The linear coefficients and noise variance are estimated pixel-wise using Neumaier and Schneider’s algorithm. Together with mean intensity they comprise a parametric space in which classification (maximum a posteriori or K-nearest neighbour) of each pixel is performed. We emphasize the novelty of studying dynamics rather than static features such as intensity in the segmentation of various structures.

Results

We assessed the automatic segmentations of ten US sequences using Dice Similarity Coefficients. The algorithm’s capability of vessel extraction was tested on three sequences where Doppler US failed to capture vasculature.

Conclusion

The modelling of image dynamics with AR process combined with MAP classifier produced robust segmentation results indicating that the method has a good potential for intra-operative use.     

Statistical shape and texture model of quadrature phase information for prostate segmentation

Purpose  

Prostate volume estimation from segmentation of transrectal ultrasound (TRUS) images aids in diagnosis and treatment of prostate hypertrophy and cancer. Computer-aided accurate and computationally efficient prostate segmentation in TRUS images is a challenging task, owing to low signal-to-noise ratio, speckle noise, calcifications, and heterogeneous intensity distribution in the prostate region.     

Combinative multi-scale level set framework for echocardiographic image segmentation

In the automatic segmentation of echocardiographic images, a priori shape knowledge has been used to compensate for poor features in ultrasound images. This shape knowledge is often learned via an off-line training process, which requires tedious human effort and is highly expertise-dependent. More importantly, a learned shape template can only be used to segment a specific class of images with similar boundary shape. In this paper, we present a multi-scale level set framework for segmentation of endocardial boundaries at each frame in a multiframe echocardiographic image sequence. We point out that the intensity distribution of an ultrasound image at a very coarse scale can be approximately modeled by Gaussian. Then we combine region homogeneity and edge features in a level set approach to extract boundaries automatically at this coarse scale. At finer scale levels, these coarse boundaries are used to both initialize boundary detection and serve as an external constraint to guide contour evolution. This constraint functions similar to a traditional shape prior. Experimental results validate this combinative framework.     

Learning to estimate out-of-plane motion in ultrasound imagery of real tissue

In freehand 3D ultrasound (US), the relative positions and orientations of the 2D US images are usually obtained from a position tracking device, at the expense of clinical convenience. As an alternative or complement to this approach, transducer motion can be inferred from image content, using image registration techniques to recover in-plane motion and speckle decorrelation to recover out-of-plane motion. One difficulty with the speckle decorrelation approach is that for real tissues, the rate of speckle decorrelation is not only transducer dependent, but also medium dependent. This paper proposes a novel method for estimating the elevational correlation length of US signals in such media by learning its relationship to in-plane image statistics from a pool of synthetic US imagery generated from virtual phantoms of varied micro-structure. Learning takes place within a sparse Gaussian process regression framework. In experiments with synthetic US imagery and real imagery of animal tissue, the approach is shown to generalise well across transducer and medium changes, with performance better than a method based on speckle classification and comparable to our implementation of the heuristic state-of-the-art method. The proposed approach better lends itself to improvement through the creation of more realistic training sets.     

A comparison of two algorithms for automated stone detection in clinical B-mode ultrasound images of the abdomen

Ultrasound (US) imaging is an indispensible technique for detection of abdominal stones which are a serious health hazard. Segmentation of stones from abdominal ultrasound images presents a unique challenge because these images contain strong speckle noise and attenuated artifacts. In clinical situations where a large number of stones must be identified, traditional methods such as manual identification become tedious and lack reproducibility too. The necessity of obtaining high reproducibility and the need to increase efficiency motivates the development of automated and fast procedures that segment out stones of all sizes and shapes in medical images by applying image segmentation techniques. In this paper we present and compare two fully automatic and unsupervised methods for robust stone detection in B-mode ultrasound images of the abdomen. Our approaches are based on the marker controlled watershed segmentation, along with some pre-processing and post-processing procedures that eliminate the inherent problems associated with medical ultrasound images. The first algorithm (Algorithm I) utilizes the advantage of the Speckle reducing anisotropic diffusion (SRAD) technique, along with unsharp filtering and histo- gram equalization for removal of speckle noise, and the second algorithm (Algorithm II) is based on the log decompression model which too serves as a tool for minimization of speckle. Experimental results obtained from processing a set of 50 ultrasound images ensure the robustness of both the proposed algorithms. Comparative results of both the algorithms based on efficiency and relative error in stone area have been provided.     

Visual enhancement of digital ultrasound images: wavelet versus Gauss–Laplace contrast pyramid

Purpose Noise is the principal factor which hampers the visual quality of ultrasound images, sometimes leading to misdiagnosis. Speckle noise in ultrasound images can be modeled as a random multiplicative process. Speckle reduction techniques were applied to digital ultrasound images to suppress noise and improve visual quality. Rationale Previous reports indicate that wavelet filtering performs best for speckle reduction in digital ultrasound images. Reportes on x-ray images compared wavelet filtering with Laplace-Gauss contrast enhancement (LGCE) showed that the LCGE performed better. As LGCE was never been applied to Ultrasound images, this study compared two filtering approaches for speckle reduction on digital ultrasound images. Methods Two methods were implemented and compared. The first method uses the wavelet soft threshold (WST) approach for enhancement. The second method is based on multiscale Laplacian-Gaussian contrast enhancement (LGCE). LGCE is derived from the combination of a Gaussian pyramid and a Laplacian one. Contrast enhancement is applied on local scale by using varying sizes of median filter. Results The two methods were applied to synthetic and real ultrasound images. A comparison between WST and LGCE methods was performed based on noise level, artifacts and subjective image quality. Conclusion WST visual enhancement provided better results than LGCE for selected ultrasound images.     

Segmentation of breast ultrasound images based on active contours using neutrosophic theory

Purpose

Ultrasound imaging is an effective approach for diagnosing breast cancer, but it is highly operator-dependent. Recent advances in computer-aided diagnosis have suggested that it can assist physicians in diagnosis. Definition of the region of interest before computer analysis is still needed. Since manual outlining of the tumor contour is tedious and time-consuming for a physician, developing an automatic segmentation method is important for clinical application.

Methods

The present paper represents a novel method to segment breast ultrasound images. It utilizes a combination of region-based active contour and neutrosophic theory to overcome the natural properties of ultrasound images including speckle noise and tissue-related textures. First, due to inherent speckle noise and low contrast of these images, we have utilized a non-local means filter and fuzzy logic method for denoising and image enhancement, respectively. This paper presents an improved weighted region-scalable active contour to segment breast ultrasound images using a new feature derived from neutrosophic theory.

Results

This method has been applied to 36 breast ultrasound images. It generates true-positive and false-positive results, and similarity of 95%, 6%, and 90%, respectively.

Conclusion

The purposed method indicates clear advantages over other conventional methods of active contour segmentation, i.e., region-scalable fitting energy and weighted region-scalable fitting energy.

     

Simulation model for assessing quality of ultrasound strain estimation in abdominal aortic aneurysm

The purpose of this study was to develop a simulation model for evaluating methods for ultrasound strain estimation in abdominal aortic aneurysms. Wall geometry was obtained from a real ultrasound image and wall motion was simulated applying realistic blood pressures to a nonlinear viscoelastic wall model. The ultrasound simulation included speckle, absorption and angle dependent reflection. Gaussian white noise was added to simulate various noise levels. Despite not fully replicating real ultrasound images, the model simulated realistic circumferential variations in intensity and realistic speckle patterns and has potential for initial evaluation of strain estimation methods.     

Self-co-attention neural network for anatomy segmentation in whole breast ultrasound

The automated whole breast ultrasound (AWBUS) is a new breast imaging technique that can depict the whole breast anatomy. To facilitate the reading of AWBUS images and support the breast density estimation, an automatic breast anatomy segmentation method for AWBUS images is proposed in this study. The problem at hand is quite challenging as it needs to address issues of low image quality, ill-defined boundary, large anatomical variation, etc. To address these issues, a new deep learning encoder-decoder segmentation method based on a self-co-attention mechanism is developed. The self-attention mechanism is comprised of spatial and channel attention module (SC) and embedded in the ResNeXt (i.e., Res-SC) block in the encoder path. A non-local context block (NCB) is further incorporated to augment the learning of high-level contextual cues. The decoder path of the proposed method is equipped with the weighted up-sampling block (WUB) to attain class-specific better up-sampling effect. Meanwhile, the co-attention mechanism is also developed to improve the segmentation coherence among two consecutive slices. Extensive experiments are conducted with comparison to several the state-of-the-art deep learning segmentation methods. The experimental results corroborate the effectiveness of the proposed method on the difficult breast anatomy segmentation problem on AWBUS images.     

Registration of 3D fetal neurosonography and MRI

We propose a method for registration of 3D fetal brain ultrasound with a reconstructed magnetic resonance fetal brain volume. This method, for the first time, allows the alignment of models of the fetal brain built from magnetic resonance images with 3D fetal brain ultrasound, opening possibilities to develop new, prior information based image analysis methods for 3D fetal neurosonography. The reconstructed magnetic resonance volume is first segmented using a probabilistic atlas and a pseudo ultrasound image volume is simulated from the segmentation. This pseudo ultrasound image is then affinely aligned with clinical ultrasound fetal brain volumes using a robust block-matching approach that can deal with intensity artefacts and missing features in the ultrasound images. A qualitative and quantitative evaluation demonstrates good performance of the method for our application, in comparison with other tested approaches. The intensity average of 27 ultrasound images co-aligned with the pseudo ultrasound template shows good correlation with anatomy of the fetal brain as seen in the reconstructed magnetic resonance image.     

A coupled deformable model for tracking myocardial borders from real-time echocardiography using an incompressibility constraint

Real-time three-dimensional (RT3D) echocardiography is a new image acquisition technique that allows instantaneous acquisition of volumetric images for quantitative assessment of cardiac morphology and function. To quantify many important diagnostic parameters, such as ventricular volume, ejection fraction, and cardiac output, an automatic algorithm to delineate the left ventricle (LV) from RT3D echocardiographic images is essential. While a number of efforts have been made towards segmentation of the LV endocardial (ENDO) boundaries, the segmentation of epicardial (EPI) boundaries remains problematic. In this paper, we present a coupled deformable model that addresses this problem. The idea behind our method is that the volume of the myocardium is close to being constant during a cardiac cycle and our model uses this coupling as an important constraint. We employ two surfaces, each driven by the image-derived information that takes into account ultrasound physics by modeling the speckle statistics using the Nakagami distribution while maintaining the coupling. By simultaneously evolving two surfaces, the final segmentation of the myocardium is thus achieved. Results from 80 sets of synthetic data and 286 sets of real canine data were evaluated against the ground truth and against outlines from three independent observers, respectively. We show that results obtained with our incompressibility constraint were more accurate than those obtained without constraint or with a wall thickness constraint, and were comparable to those from manual segmentation.     

Modeling envelope statistics of blood and myocardium for segmentation of echocardiographic images

The objective of this study was to investigate the use of speckle statistics as a preprocessing step for segmentation of the myocardium in echocardiographic images. Three-dimensional (3D) and biplane image sequences of the left ventricle of two healthy children and one dog (beagle) were acquired. Pixel-based speckle statistics of manually segmented blood and myocardial regions were investigated by fitting various probability density functions (pdf). The statistics of heart muscle and blood could both be optimally modeled by a K-pdf or Gamma-pdf (Kolmogorov-Smirnov goodness-of-fit test). Scale and shape parameters of both distributions could differentiate between blood and myocardium. Local estimation of these parameters was used to obtain parametric images, where window size was related to speckle size (5 x 2 speckles). Moment-based and maximum-likelihood estimators were used. Scale parameters were still able to differentiate blood from myocardium; however, smoothing of edges of anatomical structures occurred. Estimation of the shape parameter required a larger window size, leading to unacceptable blurring. Using these parameters as an input for segmentation resulted in unreliable segmentation. Adaptive mean squares filtering was then introduced using the moment-based scale parameter (sigma(2)/mu) of the Gamma-pdf to automatically steer the two-dimensional (2D) local filtering process. This method adequately preserved sharpness of the edges. In conclusion, a trade-off between preservation of sharpness of edges and goodness-of-fit when estimating local shape and scale parameters is evident for parametric images. For this reason, adaptive filtering outperforms parametric imaging for the segmentation of echocardiographic images.     

Discriminant snakes for 3D reconstruction of anatomical organs

In this work a new statistic deformable model for 3D segmentation of anatomical organs in medical images is proposed. A statistic discriminant snake performs a supervised learning of the object boundary in an image slice to segment the next slice of the image sequence. Each part of the object boundary is projected in a feature space generated by a bank of Gaussian filters. Then, clusters corresponding to different boundary pieces are constructed by means of linear discriminant analysis. Finally, a parametric classifier is generated from each contour in the image slice and embodied into the snake energy-minimization process to guide the snake deformation in the next image slice. The discriminant snake selects and classifies image features by the parametric classifier and deforms to minimize the dissimilarity between the learned and found image features. The new approach is of particular interest for segmenting 3D images with anisotropic spatial resolution, and for tracking temporal image sequences. In particular, several anatomical organs from different imaging modalities are segmented and the results compared to expert tracings.