Wavelet Filtering

The term image enhancement is also a very general term, which encompasses many techniques. We have recognized that it is often used to describe the outcome of filtering. If the definition of enhancement is applying a process to the image that results in a better overall image appearance, then the term is a misnomer. Linear filtering blocks a portion of the true signal in most applications, which is probably not best defined as enhancement. In this section we provide a qualitative description of...

Multimodal Techniques

Comparisons of datasets obtained from individual subjects between imaging modalities are very important for the evaluation of the normal physiologic responses of the anatomic structure or the pathophysiological changes that accompany disease states. Likewise, it is also critical to compare data between individuals both within and across different imaging modalities. Unfortunately, many structures of interest, particularly in the brain, are often smaller than the spatial resolution of the...

Edge Based Segmentation

Edge-based segmentation approaches have two major components (1) edge detection and (2) edge linking following to determine the edges and the regions. Loosely speaking, an edge is a collection of connected pixels that lie on the boundary between two homogeneous regions having different intensities. Therefore, edges can be defined as abrupt changes in pixel intensity that can be reflected by the gradient information. A number of edge detectors have been defined based on the first-order or...

CADiagnosis Algorithm Design

The diagram in Fig. 13.1 shows the major components of a CADiagnosis algorithm aiming at the differentiation between benign and malignant lesions. Based on this diagram, one may distinguish two major pathways to algorithm development 1. In one approach, a fully automated scheme is developed. Namely, the algorithm includes automated detection, feature selection, and classification modules. In this case, the diagnosis component of the algorithm may be considered as preceded by a CADetection...

Segmentation Based on Anatomical Models

During the past few years, the usage of deformable anatomical atlases has been extensively investigated as an appealing tool for the coding of prior anatomical information for image interpretation. The method is based on a representative deterministic 14 or probabilistic 15 image volume as an anatomical model. For this the actual patient data has to be spatially normalized, thus it has to be mapped onto the template that conforms to the standard anatomical space used by the model. The applied...

Processing Image Segmentation

Automatic segmentation of CT images admittedly presents significant challenges in computer vision 42 . The primary reason is that the organs are flexible and their size and shape varies as a function of patient characteristics and imaging parameters. Organs are usually accurately localized on CT slices (Fig. 4.4.) but the detection and separation of their boundaries from those of their neighbors and the background is often a difficult task due to the obscure, fuzzy, and irregular edges that are...

Knowledge Representation by Image Grouping on Various Criteria

This approach to implementing a knowledge-based component attempts to adap-tively determine optimum parameter settings for groups of images on the basis of image feature vectors. The feature vector is used to group images accordingly, that in turn serve as a form of a priori knowledge for use in subsequent components. In this way components may be trained to operate on particular image groupings with different parameter settings. In mammography, Zheng et al. 9 propose an adaptive computer-aided...

Mammographic Cluster Classification Single View Application

A set of 260 single-view mammograms with calcification clusters was first used for the validation of the CADiagnosis algorithm described previously. The set included 138 calcification clusters associated with benign disease that are commonly referred to as benign calcifications or clusters and 122 calcification clusters associated with cancer that are commonly referred to as malignant calcifications or clusters. All mammograms were selected from the patient files of the H. Lee Moffitt Cancer...

Novel Algorithm for Pancreatic Tumor Detection and Classification

Fuzzy-based segmentation and classification techniques have been used in various medical imaging applications although not pancreatic cancer 46-48 . An application closest to CT pancreatic imaging with analogous problems is the magnetic resonance imaging (MRI) of the brain and brain tumors. Unsupervised, supervised, and semisupervised fuzzy c-means (FCM) algorithms and knowledge-guided FCM segmentation have been successfully applied to brain tumor MRI applications 49-55 . Similar approaches...

Polynomial Contrast Enhancement

Let f E T be a gray-level image with T n, , imax) C R aset of rational numbers. Let U Wmin, , umax C R be a second set of rational numbers. An application is called gray-level transformation. For convenience, the gray-level transformation is constructed in such a way that it assigns 1 (u + umax) to the mean value xt of the original image f. Instead of t and u, we consider in the following the variables t and v defined by A polynomial gray-level transformation can then be defined as follows 1 (t...

Classification Algorithm

Classification was done with a three-layer, feed-forward artificial neural network ANN consisting of an input layer, one hidden layer, and an output layer. The NevProp4 backpropagation software was used in this study. NevProp4 is a general backpropagation algorithm developed by Philip H. Goodman at the University of Nevada, Reno 60 . Figure 13.15 shows a diagram of the network structure. The feature vector of the input layer consisted of 14 elements features defined in the previous stage Table...

Ra0d min Ra0c fmc d and R Ra0d1212

We now show that, for all c, d e V, c, d is consistent. To do this, we assume that there is a c, d and an m such that R am R ac and yet none of Eqs. 12.10 - 12.12 holds and show that this leads to a contradiction. A consequence of our assumption is that c d and at least one of the 1 RIad lt min , fm c, d 12.13 1 Biad min RV0c, fm c, d and a IRIad 12.14 We may assume that IRIaC gt 0 and that m c, d gt 0, for otherwise one of Eqs. 12.11 or 12.12 clearly holds. Hence 1 RIa 1 RIa ir, for some 1 lt...

Live Wire Segmentation

The first class of algorithms reviewed, which are usually referred to as Live-Wire algorithms 38 , belongs to the field of dynamic programming. The foundations of these algorithms lie in the F algorithm 39 , and will briefly be sketched here. In a Live-Wire algorithm, the image I is considered as a discrete neighborhood graph, where each pixel corresponds to a node in the graph. Generally an 8-Neighborhood N8, Moore neighborhood is defined, so that diagonal connections are allowed. A cost...

Helical CT Imaging Characteristics

CT was developed in the early 1960s with the first clinical system installed in 1971 26 . Today, there are five generations of CT scanners characterized by different scanning conditions and properties. Helical or spiral CT is the latest generation of scanners that combines a continuous rotation of the X-ray source and ring of detectors with a continuous movement of the examination table. Hence, data are acquired continuously while the patient is moved through the gantry 27, 28 . Figure 4.3...

Retinal Image Segmentation from Stereo Fundus Images

Objects such as blood vessels, optic disk, and optic cup in retinal images are crucial in monitoring and detecting the progression of retinal diseases such as vascular diseases, glaucoma hypertension, and diabetic retinopathy. b First pass segmentation background b First pass segmentation background c First pass segmentation CSF and other d First pass segmentation gray matter e First pass segmentation white matter f White matter mask Figure 6.12 DA segmentation of MS lesions from Fig. 6.11 a ....

Bibliography

1 Jemal, A., Thomas, A., and Murray, T., Cancer statistics, 2002, CA Cancer J. Clin., Vol. 52, pp. 23-47, 2002. 2 Kern, S., Tempero, M., and Conley, B., Co-Chairs , Pancreatic cancer An agenda for action, Report of the Pancreatic Cancer Progress Group, National Cancer Institute, February 2001. 3 Kuvshinoff, B. W. and Bryer, M. P., Treatment of resectable and locally advanced pancreatic cancer, Cancer Control, Vol. 7, No. 5, pp. 428-436, 2000. 4 Lin, Y., Tamakoshi, A., Kawamura, T., Inaba, Y.,...

Introduction

Vascular disease, stroke, and arterial dissection or rupture of coronary arteries are considered some of the main causes of mortality in present days. The behavior of the atherosclerotic lesions depends not only on the degree of lumen narrowing but also on the histological composition that causes that narrowing. Therefore, tissue characterization is a fundamental tool for studying and diagnosing the pathologies and lesions associated to the vascular tree. Although important, tissue...

Pelvic Bone Tumor and Cortex Visualization from 3D CT Data

A single-scale sheet filter was applied to pelvic CT images for bone cortex enhancement. The purpose was to visualize the distribution of bone tumors and localize them in relation to the pelvic structure for biopsy planning as well as diagnosis 32 . Healthy bone cortex tissues and bone tumors have similar original CT values. However, bone cortices are sheet-like in structure, while tumors are not. Thus, enhanced bone cortices using sheet enhancement filtering are expected to be discriminated...

Multiscale Computation and Integration of Filter Responses

Local structures can exist at various scales. For example, vessels and bone cortices can, respectively, be regarded as line and sheet structures with various widths. In order to make filter responses tunable to a width of interest, the derivative computation for the gradient vector and the Hessian matrix is combined with Gaussian convolution. By adjusting the standard deviation of Gaussian convolution, local structures with a specific range of widths can be enhanced. The Gaussian function is...

Using Anisotropic Gaussian Blurring Based on Voxel Anisotropy

We have assumed that Gaussian blurring combined with derivative computation is isotropic as shown in Eq. 10.51 . Another choice is to use anisotropic Gaussian blurring corresponding to voxel anisotropy, which is given by 9xx x axy, az Gauss x, y aXy Gauss z az , 10.75 where az and axy are determined so as to satisfy A-, and thus az Azaxy because we assumed Axy 1. Figure 10.20 c shows plots of measured thickness obtained using anisotropic Gaussian blurring when Az 2 and axy The plots using...

Subvoxel Edge Localization and Width Measurement

An adaptive directional second derivative is applied at each voxel based on its corresponding moving frame. The directional derivative is taken along the perpendicular from the voxel to the medial axis or surface. The zero-crossing points of the directional second derivatives are localized at subvoxel resolution to determine the precise region boundaries and quantitate the widths. At every voxel within the candidate regions, the directional second derivative is calculated depending on its...

Validating the Numerical Simulation by in Vitro Experiments

To validate the numerical simulation, the postprocessing method for thickness determination was used to measure real MR images of two different objects, an acrylic plate phantom. A phantom of sheet-like objects with known thickness was used. It consisted of four acrylic plates of 80 x 80 mm2 with thickness t 1.0,1.5,2.0, and 3.0 mm , placed parallel to each other with an interval of 30 mm Fig. 10.21 a . 3 mm T- 1.5 mm r 1 mm tC T - 2 mm S Figure 10.21 Acrylic plate phantom and its MR images. a...

Double Network Mapping Overview

The double network mapping DNM method is used to predict the target contrast enhancement using two ANNs for each enhancement method. The aim is to learn a mapping based on a set of gray-scale features FROI from a given mammogram, with a quantitative measure of segmentation performance, S. The segmentation performance is quantified following contrast enhancement, for each enhancement method m, where 1 lt m lt M from a set of M enhancement methods. The two submappings are detailed below 1....