We analyze the segmentation of sparse data using the 3D variant of Active Shape Models by van Assen et al.
(SPASM). This algorithm is designed to segment volumetric data represented by multiple planes with arbitrary
orientations and with large undersampled regions. With the help of statistical shape constraints, the complicated
interpolation of the sliced data is replaced by a mesh-based interpolation. To overcome large void areas without
image information the mesh nodes are updated using a Gaussian kernel that propagates the available information
to the void areas. Our analysis shows that the accuracy is mostly constant for a wide range of kernel scales,
but the convergence speed is not. Experiments on simulated 3D echocardiography datasets indicate that an
appropriate selection of the kernel can even double the convergence speed of the algorithm. Additionally, the
optimal value for the kernel scale seems to be mainly related to the spatial frequency of the model encoding
the statistical shape priors rather than to the sparsity of the sliced data. This suggests the possibility to precalculate
the propagation coefficients which would reduce the computational load up to 40% depending on the
spatial configuration of the input data.
KEYWORDS: Single photon emission computed tomography, 3D modeling, Image segmentation, Data modeling, Statistical modeling, Monte Carlo methods, Error analysis, Statistical analysis, Heart, Gold
Over the course of the last two decades, myocardial perfusion with Single Photon Emission Computed Tomography
(SPECT) has emerged as an established and well-validated method for assessing myocardial ischemia,
viability, and function. Gated-SPECT imaging integrates traditional perfusion information along with global
left ventricular function. Despite of these advantages, inherent limitations of SPECT imaging yield a challenging
segmentation problem, since an error of only one voxel along the chamber surface may generate a huge difference
in volume calculation. In previous works we implemented a 3-D statistical model-based algorithm for Left Ventricle
(LV) segmentation of in dynamic perfusion SPECT studies. The present work evaluates the relevance of
training a different Active Shape Model (ASM) for each frame of the gated SPECT imaging acquisition in terms
of their subsequent segmentation accuracy. Models are subsequently employed to segment the LV cavity of gated
SPECT studies of a virtual population. The evaluation is accomplished by comparing point-to-surface (P2S)
and volume errors, both against a proper Gold Standard. The dataset comprised 40 voxel phantoms (NCAT,
Johns Hopkins, University of of North Carolina). Monte-Carlo simulations were generated with SIMIND (Lund
University) and reconstructed to tomographic slices with ASPIRE (University of Michigan).
KEYWORDS: 3D modeling, Image segmentation, 3D image processing, Ultrasonography, Transducers, Tissues, In vivo imaging, Data modeling, Point spread functions, Image resolution
In this paper a study of 3D ultrasound cardiac segmentation using Active Shape Models (ASM) is presented.
The proposed approach is based on a combination of a point distribution model constructed from a multitude of
high resolution MRI scans and the appearance model obtained from simulated 3D ultrasound images. Usually
the appearance model is learnt from a set of landmarked images. The significant level of noise, the low resolution
of 3D ultrasound images (3D US) and the frequent failure to capture the complete wall of the left ventricle (LV)
makes automatic or manual landmarking difficult. One possible solution is to use artificially simulated 3D US
images since the generated images will match exactly the shape in question. In this way, by varying simulation
parameters and generating corresponding images, it is possible to obtain a training set where the image matches
the shape exactly. In this work the simulation of ultrasound images is performed by a convolutional approach.
The evaluation of segmentation accuracy is performed on both simulated and in vivo images. The results obtained
on 567 simulated images had an average error of 1.9 mm (1.73 ± 0.05 mm for epicardium and 2 ± 0.07 mm for
endocardium, with 95% confidence) with voxel size being 1.1 × 1.1 × 0.7 mm. The error on 20 in vivo data was
3.5 mm (3.44 ± 0.4 mm for epicardium and 3.73 ± 0.4 mm for endocardium). In most images the model was
able to approximate the borders of myocardium even when the latter was indistinguishable from the surrounding
tissues.
As it is known, the impulsive noise appears on the image in the form of randomly distributed pixels of random brightness. Impulses themselves usually differ much from the surrounding pixels in brightness. The main topic of the paper is the introduction of the new impulse detection criteria, and their application to such filters as median, rank-order and cellular neural Boolean. Three impulse detectors are considered. The Rank Impulse Detector uses such property of impulse that its rank in variation series is usually quite different from rank of the median. Exponential Median Detector uses the exponent of the difference between the local median and the value of pixel to detect the impulse. Combination of these two detectors forms the Enhanced Rank Impulse Detector and integrates advantages of both of them. In combination with filter it allows iterative filtering without further image destruction.
There are different techniques available for solving of the restoration problem including Fourier domain techniques, regularization methods, recursive and iterative filters to name a few. But without knowing at least approximate parameters of the blur, these methods often show poor results. If incorrect blur model is chosen then the image will be rather distorted much more than restored. The original solution of the blur and blur parameters identification problem is presented in this paper. A neural network based on multi-valued neurons is used for the blur and blur parameters identification. It is shown that it is possible to identify the type of the distorting operator by using simple single-layered neural network. Four types of blur operators are considered: defocus, rectangular, motion, and Gaussian ones. The parameters of the corresponding operator are identified by using a similar neural network. After identification of the blur type and its parameters the image can be restored using different methods. Some fundamentals of image restoration techniques are also considered.
As a rule, blur is a form of bandwidth reduction of an ideal image owing to the imperfect image formation process. It can be caused by relative motion between the camera and the original scene, or by an optical system that is out of focus. Today there are different techniques available for solving of the restoration problem including Fourier domain techniques, regularization methods, recursive and iterative filters to name a few. But without knowing at least approximate parameters of the blur, these filters show poor results. If incorrect blur model is chosen then the image will be rather distorted much more than restored. The original solution of the blur and blur parameters identification problem is presented in this paper. A neural network based on multi-valued neurons is used for the blur and blur parameters identification. It is shown that using simple single-layered neural network it is possible to identify the type of the distorting operator. Four types of blur are considered: defocus, rectangular, motion and Gaussian ones. The parameters of the corresponding operator are identified using a similar neural network. After a type of blur and its parameters identification the image can be restored using several kinds of methods. Some fundamentals of image restoration are also considered.
Removal of periodic and quasi-periodic patterns from photographs is an important problem. There are a lot of sources of this periodic noise, e.g. the resolution of the scanner used to scan the image affects the high frequency noise pattern in the acquired image and can produce moire patterns. It is also characteristic of gray scale images obtained from single-chip video cameras. Usually periodic and quasi-periodic noise results peaks in image spectrum amplitude. Considering this, processing in the frequency domain is a much better solution than spatial domain operations (blurring for example, which can hide the periodic patterns at the cost of the edge sharpness reduction). A new frequency domain filter for periodic and quasi-periodic noise reduction is introduced in this paper. This filter analyzes the image spectrum amplitude using a local window, checks every spectral coefficient whether it needs the filtering and if so, replaces it with the median taken from the local window. To detect the peaks in the spectrum amplitude, a ratio of the current amplitude value to median value is used. It is shown that this ratio is stable for the non-corrupted spectral coefficients independently of the frequencies they correspond to. So it is invariant to the position of the peaks in the spectrum amplitude. This kind of filtering completely eliminates periodic noise, and shows quite good results on quasi-periodic noise while completely preserves the image boundaries.
Multi-valued neurons (MVN) are the neural processing elements with complex-valued weights and high functionality. It is possible to implement an arbitrary mapping described by partial-defined multiple-valued function on the single MVN. The MVN-based neural networks are applied to temporal classification of images of gene expression patterns, obtained by confocal scanning microscopy. The classification results confirmed the efficiency of this method for image recognition. It was shown that frequency domain of the representation of gene expression images is highly effective for their description.
Some important ideas of image recognition using neural network based on multi-valued neurons are being developed in this paper. We are going to discuss the recognition of color images, which is reduced to recognition of gray-scale images. An approach, which has been developed, is illustrated by simulation results. Recognition of distortion (blur) types, distortion parameters and recognition of images with distorted training set using the same neural network is also considered. At this time Gaussian blur and motion blur were taken as distortions. This part of work is also illustrated by simulation results.
Nonlinear cellular neural filters (NCNF) were introduced recently. They are based on the complex non-linearity of multi-valued and universal binary neurons. NCNF include multi- valued filters and cellular neural Boolean filters. Applications of the NCNF to noise reduction, extraction of image details and precise edge detection have been considered recently. This paper develops the previous ideas and presents the new results. The following problems are considered in the paper: (1) Solution of the Super-resolution problem using iterative extrapolation of the orthogonal spectra and final correction of the resulting image using NCNF; (2) Precise edge detection using NCNF within a 5 X 5 window and precise edge detection for the color images.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.