Image Processing Ebooks Catalog

Learn Photo Editing

This online course gives professional advice and instructions for how to photoshop pictures for any purpose that you could need them for. If you need to retouch your portraits, this gives you the tools to edit the image so that your model is sure to be happy with the results. If you need to create cartoon characters, you can learn how to do that in a very short amount of time. You can even learn the more advanced skills, like how to make facial features stand out in the picture without having to retouch the photo. You can learn how to take your normal photos and turn them into glossy, high resolution advertisements. Whatever skills you want to learn, and whatever application your photos will be needed in, this course can give you the tools that you need in order to create the most beautiful photoshoot that you've ever done. Read more...

Learn Photo Editing Summary


4.8 stars out of 26 votes

Contents: Premium Membership
Author: Patrick
Official Website:
Price: $27.00

Access Now

My Learn Photo Editing Review

Highly Recommended

I usually find books written on this category hard to understand and full of jargon. But the author was capable of presenting advanced techniques in an extremely easy to understand language.

In addition to being effective and its great ease of use, this eBook makes worth every penny of its price.

D GVF Deformable Models and Results

FIGURE 6 (a) A 160 x 160-pixel magnetic resonance image of the left ventricle of a human heart (b) the edge map V(Ga * I) with a 2.5 (c) the GVF field (shown subsampled by a factor of 2) and (d) convergence of the GVF deformable contour. Reprinted from C. Xu and J. L. Prince, Snakes, shapes, and gradient vector flow. IEEE Trans, on Image Processing, 7(3) 359 369, March, 1998. 1998 IEEE. FIGURE 6 (a) A 160 x 160-pixel magnetic resonance image of the left ventricle of a human heart (b) the edge map V(Ga * I) with a 2.5 (c) the GVF field (shown subsampled by a factor of 2) and (d) convergence of the GVF deformable contour. Reprinted from C. Xu and J. L. Prince, Snakes, shapes, and gradient vector flow. IEEE Trans, on Image Processing, 7(3) 359 369, March, 1998. 1998 IEEE.

Genetic Algorithm Approach

The GA starts from a population of randomly selected chromosomes, which are represented by binary or gray-level digital strings. Each chromosome consists of a number of genes (bits in the string) and corresponds to a possible solution of the problem. For feature selection in medical image processing, a chromosome is typically represented by a binary coded feature string, with 1 indicating the presence of a gene (the feature is used in the classifier) and 0 indicating its absence (the feature is not used in the classifier). (2) Evaluation. In this step a fitness function is applied to evaluate the fitness of all chromosomes in the population. The type of fitness function or criterion is determined by the specific applications. Since the main purpose of medical image processing is to improve the diagnostic accuracy, and ROC methodology has become a standard to evaluate the diagnostic accuracy, the areas under ROC curve (Az) are often used as a fitness criterion....

Image Standardization in PACS

Standardization facilitates the medical requirements and use of the data. In this approach we concentrate on the image content adjustment to make images more readable and of better quality in preparation for medical diagnosis. This step also makes advanced image processing phase easier and permits some preprocessing steps not to be considered at the level of development of the methodology, which results in a computer added diagnosis. Although this group of standardization procedures is related to image information itself without referring to image format or intersystem communication, standardization functions are integrated with clinical PACS and installed at various sites of the system. Various definitions of background have been introduced already. Usually it is described as an area of no importance attached to a region that is to be enhanced. Furthermore, very often it affects the visual image quality. In image standardization three various areas may be referred to as background....

Comparison Between Artificial Neural Networks and Bayesian Belief Networks

In medical image processing, usually a large number of features are extracted. In a complex and multidimensional Because of the black box approach and possible data overfitting in training an ANN, some researchers found that it might be difficult to convince the physicians to accept and act on advice from an ANN-based computer-assisted system 24 . In contrast, a BBN can explain the reasoning process and offer an efficient and principled approach for avoiding data over-fitting 7 . Thus, some investigators suggested that a BBN should be more reliable than an ANN used as a computerassisted tool for diagnosis of medical images 12 . However, there was no experimental evidence to support that a BBN is better than an ANN in medical image processing. In fact, an important principle in pattern classification indicates that feature extraction is the most crucial aspect of classification. If features are not capable of discriminating classes of interest at all, the resulting recognition...

Statistical Methods for Brain Segmentation

Kapur et al. 19 segment the brain in 3D gradient-echo MR images by combining the statistical classification of Wells et al. 34 with image processing methods. A single channel, non-parametric, multiclass implementation of Well's classifier based on tissue type training points is used to classify brain tissues. Further morphological processing is needed to remove connected nonbrain components. The established brain contours are refined using an algorithm based on snakes. The combination of statistical classification of brain tissue, followed by morphological operators, is effective in segmenting the brain from other structures such as orbits in a semiautomated fashion. Furthermore, Wells's statistical classification method also reduces the effect of RF inhomogeneity. However, Kapur's method requires some interaction to provide tissue training pixels, and in 10 of volumes studied interaction is needed to remove nonconnected brain tissue. The method is computationally intensive and has...

Image Segmentation and Manipulation

IMPROMPTU (IMage PROcessing Module for Prototyping, Testing, and Utilizing image-analysis processes) provides a graphical user interface system for constructing, testing, and executing automatic image analysis processes. Elaborate image analyses can be performed by constructing a sequence of simpler image processing and analysis functions such as filters, edge detectors, and morphological operators. The interface currently links to a library (VIPLIB Volumetric Image Processing function LIBrary) of 1D, 2D, and 3D image processing and analysis functions developed at Pennsylvania State University. These scripts, used in conjunction with VIDA's segmentation modules, can automate a series of processes that need to be applied to several data sets. Users can create and add customized functions to the library.

Adaptive Wiener Filters

The standard formulation of the Wiener filter has met limited success in image processing because of its lowpass characteristics, which give rise to unacceptable blurring of lines and edges. If the signal is a realization of a non-Gaussian process such as in natural images, the Wiener filter is outperformed by nonlinear estimators. One reason why the Wiener filter blurs the image significantly is that a fixed filter is used throughout the entire image, the filter is space invariant.

Two Dimensional Extension

For image processing applications, the ID structures discussed previously are simply extended to two dimensions. We first adopt the method proposed by Mallat and Zhong 20 , shown in Fig 11, where filter I (m) 1 + h (m) 2, and h (m), k (m), and S (m) are the same filters as constructed for the 1D case. The left side of Fig. 10 corresponds to analysis (decomposition) while the right side is synthesis (reconstruction). The bar above some of the analytic forms of the synthesis filters refers to the complex conjugate.

Artificial Neural Networks

An ANN is a computational structure inspired by the study of biological neural processing. It has some capability for simulating the human brain and learning or recognizing a pattern based on partial (incomplete) information. Although there are many types of ANN topologies (e.g., from a relatively simple perceptron to a very complex recurrent network), by far the most popular network architecture is a multilayer feedforward ANN trained using the back-propagation method. Because it has been proved that only one hidden layer is enough to approximate any continuous function 10 , most ANNs used in medical image processing are three-layer networks. In this chapter, we only discuss three-layer feedforward ANNs, but the conclusion also should be applicable to other types of ANNs. Once the topology of an ANN (the number of neurons in each layer) is decided, the ANN needs to be trained either in supervised or unsupervised mode using a set of training samples. In the supervised training the...

Effect of Validation Methods

However, selecting a large training database in medical image processing is not an easy task and it may be infeasible in many applications. In reality, the size of databases used in many studies reported to date is very limited. Thus, different cross-validation methods have been widely used to evaluate the performance of an ANN or a BBN. Although there are many theoretically sound techniques for validating computerassisted diagnosis or classification schemes for medical images 6 , most are based on the assumption that the training database covers the entire sample space sufficiently well. When the case domain is adequately sampled, and the investigator takes great care not to overtrain the classifier, these are valid approaches. This is typically the case when the feature domain is reasonably limited and well defined in many other fields, such as recognition of optical characters or mechanical parts in an assembly line. Unfortunately, it is not the case in many clinical applications....

Illustrative Visualization

Two-dimensional (2D) image processing techniques and displays formed the second generation systems. Some of the earliest 2D visualization tools were image processing techniques for enhancing image features that otherwise may have been ignored. Feature extraction techniques, expert systems, and neural networks applications were developed along with second generation visualization systems.

Terminology and Other Pitfalls

Since the need for interpolation arises so often in practice, its droning use makes it look like a simple operation. This is deceiving, and the fact that the interpolation terminology is so perplexing hints at hidden difficulties. Let us attempt to befuddle the reader The cubic B-spline is a piecewise polynomial function of degree 3. It does not correspond to what is generally understood as cubic convolution, the latter being a Keys' function made of piecewise polynomials of degree 3 (like the cubic B-spline) and of maximal order 3 (contrary to the cubic B-spline, for which the order is 4). No B-spline of sufficient degree should ever be used as an interpolant int, but a high-degree B spline makes for a high-quality synthesis function The Appledorn function of degree 4 is no polynomial and has order zero. There is no degree that would be associated to the sinc function, but its order is infinite. Any polynomial of a given degree can be represented by splines of the same degree, but,...

Relationships Between Quality Measures

In previous sections we have studied how certain parameters such as percent measurement error and subjective scores appear to change with bit rate. It is assumed that bit rate has a direct effect on the likely measurement error or subjective score and therefore the variables are correlated. In this sense, bit rate can also be viewed as a predictor. For instance, a low bit rate of 0.36 bits per pixel (bpp) may predict a high percent measurement error or a low subjective score. If the goal is to produce images that lead to low measurement error, parameters that are good predictors of measurement error are useful for evaluating images as well as for evaluating the effect of image processing techniques. A good predictor is a combination of an algorithm and predictor variable that estimates the measurement error within a narrow confidence interval.

Measuring Information

According to Studholme 15 , it is useful to think of the registration process as trying to align the shared information between the images If structures are shared between the two images and the images are mis-aligned, then in the combined image, these structures will be duplicated. For example, when a transaxial slice through the head is mis-aligned, there may be four eyes and four ears. As the images are brought into alignment, the duplication of features is reduced and the combined image is simplified. Using this concept, registration can be thought of as reducing the amount of information in the combined image, which suggests the use of a measure of information as a registration metric. The most commonly used measure of information in signal and image processing is the Shannon-Weiner entropy measure H, originally developed as part of communication theory in the 1940s 16

Geometric Calculation of Changes in Cardiac Volumes

The relatively large size of the left ventricle and the need for global contouring of the ventricular cavity for left ventricular volume and function evaluation facilitates the use of digital image processing techniques that enhance the image quality and lead to improved edge detection accuracy 12 . When using contrast, such as in contrast ventriculograms, the background or overlapping structure can be suppressed by subtracting a mask image from the corresponding ECG-gated, opacified image. Random noise such as quantum noise can also be reduced if images from the same point in several cardiac cycles are averaged 13 . After completion of the edge detection procedure, the left ventricular boundary contours can then be used to generate left ventricular volume and regional wall motion data.

Data Acquisition for Vascular Morphometry

We are primarily concerned with imaging and image processing methods for quantifying arterial tree morphometry. As mentioned previously, for many decades after 1895 planar X-rays recorded on film were the predominant method available to image the vasculature or any other type of structure. Today, the inherently planar methods available to clinicians and researchers for studying vascular structure and disorders fall into the two broad categories of radiography (including mobile units in the clinical setting) and fluoroscopy. Fluoroscopic methods include the highly specialized and sophisticated variants employed in angiography suites and cardiac catheterization laboratories. Whereas radiographic methods are static in nature, fluoroscopic methods permit dynamic image acquisition (15 to 60 frames per second) and are therefore useful for freezing the motion of structures such as the beating heart in the interest of extracting accurate quantitative measurements. Arteriography is a term...

Computational Neuroanatomy Using Shape Transformations

The explosive growth of modern tomographic imaging methods has provided clinicians and scientists with the unique opportunity to study the structural and functional organization of the human brain, and to better understand how this organization is disturbed in many neurological diseases. Although the quest for understanding the anatomy and function of the brain has been very old, it has previously relied primarily on qualitative descriptions. The development of modern methods for image processing and analysis during the past 15 years has brought great promise for describing brain anatomy and function in quantitative ways, and for being able to characterize subtle yet important deviations from the norm, which might be associated with or lead to various kinds of diseases or disorders. Various methods for quantitative medical image analysis seem to be converging to the foundation of the emerging field of computational neuroanatomy, or more generally, computational anatomy.

Challenges in 3D Brain Imaging

Image registration is central to many of the challenges in brain imaging today. Initially developed as an image processing subspecialty to geometrically transform one image to match another, registration now has a vast range of applications. In this chapter, we review the registration strategies currently used

Image Segmentation Methods

In this section, we describe volumetric segmentation methods that will later be compared numerically. Some of these have been implemented in commercial packages such as Mayo Clinic's Analyze software l9 and MEDx image-processing software (Sensor Systems, Sterling, VA). Here our goal is to present a representative sample of available intensity-based segmentation methods for MR images. Greater detail is provided in describing the more recently developed techniques. For additional surveys on medical image segmentation, in particular MR image segmentation, the reader should refer to 20-24 .

GVF Deformable Contours

FIGURE 1 (a) The convergence of a deformable contour using (b) traditional potential forces, (c) shown close-up within the boundary concavity. Reprinted from C. Xu and J. L. Prince, Snakes, shapes, and gradient vector flow. IEEE Trans, on Image Processing, 7(3) 359-369, March, l998. l998 IEEE. FIGURE 1 (a) The convergence of a deformable contour using (b) traditional potential forces, (c) shown close-up within the boundary concavity. Reprinted from C. Xu and J. L. Prince, Snakes, shapes, and gradient vector flow. IEEE Trans, on Image Processing, 7(3) 359-369, March, l998. l998 IEEE. FIGURE 2 (a) The convergence of a deformable contour using (b) distance potential forces, (c) shown close-up within the boundary concavity. Reprinted from C. Xu and J. L. Prince, Snakes, shapes, and gradient vector flow. IEEE Trans, on Image Processing, 7(3) 359-369, March, 1998. 1998 IEEE. FIGURE 2 (a) The convergence of a deformable contour using (b) distance potential forces, (c) shown close-up within...

Discussion and Conclusions

Arterial tree morphometry is an important application of image processing and analysis in clinical practice and the biomedical sciences. The severity of coronary artery disease is routinely assessed in the clinic with the aid of sophisticated image processing software to quantify stenoses. Presurgical planning for vascular abnormalities such as cerebral aneurysms is facilitated by segmentation and visualization of the intracerebral vasculature. Clinical studies provide information about arterial morphology on a macro scale. On the other end of the scale continuum, histological and electron microscopic methods have a long history of providing valuable insights into the cellular makeup and ultrastructure of vessel walls, and the many forms of medial hypertrophy. Micro-CT techniques such as those developed in our laboratory and others 120,121 and micro-MR methods under development have the potential to shed further light on the mechanisms implicated in diseases such as pulmonary and...


Segmentation, separation of structures of interest from the background and from each other, is an essential analysis function for which numerous algorithms have been developed in the field of image processing. In medical imaging, automated delineation of different image components is used for analyzing anatomical structure and tissue types, spatial distribution of function and activity, and pathological regions. Segmentation can also be used as an initial step for visualization and compression. Typically, segmentation of an object is achieved either by identifying all pixels or voxels that belong to the object or by locating those that form its boundary. The former is based primarily on the intensity of pixels, but other attributes, such as texture, that can be associated with each pixel, can also be used for segmentation. Techniques that locate boundary pixels use the image gradient, which has high values at the edges of objects. Chapter 5 presents the fundamental concepts and...

Id Zo Id Jo

Combined (hybrid) strategies have also been used in many applications. Here are some examples Kapur etal. 58 present a method for segmentation of brain tissue from magnetic resonance images that combines the strengths of three techniques single-channel expectation maximization segmentation, binary mathematical morphology, and active contours models. Masutani et al. 73 segment cerebral blood vessels on MRA images using a model-based region growing, controlled by morphological information of local shape. A hybrid strategy 3 that employs image processing techniques based on anisotropic filters, thresholding, active contours, and a priori knowledge of the segmentation of the brain is discussed in Chapter 11. 4. G. J. Awcock, R. Thomas, Applied Image Processing. New York McGraw-Hill, Inc., 1996. 10. S. Beucher, Segmentation tools in mathematical morphology, SPIE, vol. 1350, Image Algebra and Morphological Image Processing, pp. 70-84, 1990. 16. K. R. Castleman, Digital Image Processing....


Interactive (semiautomatic) algorithms and fully automatic algorithms represent two alternative approaches to computerized medical image analysis. Certainly automatic interpretation of medical images is a desirable, albeit very difficult, long-term goal, since it can potentially increase the speed, accuracy, consistency, and reproducibility of the analysis. However, the interactive or semiautomatic methodology is likely to remain dominant in practice for some time to come, especially in applications where erroneous interpretations are unacceptable. Consequently, the most immediately successful deformable model based techniques will likely be those that drastically decrease the labor intensiveness of medical image processing tasks through partial automation and significantly increase their reproducibility, while still allowing for interactive guidance or editing by the medical expert. Although fully automatic techniques based on deformable models will likely not reach their full...

Philosophical Issues

In addition to the advantages that the evaluation protocol confers on the originals, physician training also provides a bias for existing techniques. Radiologists are trained in medical school and residency to interpret certain kinds of images, and when asked to look at another type of image (e.g., compressed or highlighted) they may not do as well just because they were not trained on those. Highly compressed images have lower informational content than do originals, and so even a radiologist carefully trained on those could not do as well as a physician looking at original images. But with image enhancement techniques or slightly compressed images,


The increasingly important role of medical imaging in the diagnosis and treatment of disease has opened an array of challenging problems centered on the computation of accurate geometric models of anatomic structures from medical images. Deformable models offer an attractive approach to tackling such problems, because these models are able to represent the complex shapes and broad shape variability of anatomical structures. Deformable models overcome many of the limitations of traditional low-level image processing techniques, by providing compact and analytical representations of object shape, by incorporating anatomic knowledge, and by providing interactive capabilities. The continued development and refinement of these models should remain an important area of research into the foreseeable future.

Microscopy Imaging

Application of 3D visualization and analysis techniques to the field of microscopy has grown significantly in recent years 8,17,37,40,47,64 . These techniques have been successfully applied in light and electron microscopy, but the advent of confocal microscopy and other 3D microscope modalities have led to the rapid growth of 3D visualization of microscopic structures. Light microscope images digitized directly from the microscope can provide a 3D volume image by incrementally adjusting the focal plane, usually followed by image processing to deconvolve the image, removing the blurred out-of-focus structures. Similarly, electron microscopy will generate multiple planes by controlling the plane of focus, with further processing necessary for selective focal plane reconstruction. Confocal microscopy, however, uses laser energy with precise

Feature Selection

After training and testing databases are established, the images are usually preprocessed using various techniques of filtering, segmentation, and transformation, to define the regions of interest (ROIs) in the image. Then, a computer program can be used to compute or extract features from each ROI in the processed images. Many different image features (i.e., intensity-based, geometrical, morphological, fractal dimension, and texture features) have been used in medical image processing. Feature extraction can be considered as data compression that removes irrelevant information and preserves relevant information from the raw data. However, defining effective features is a difficult task. Besides the complex and noisy nature of medical images, another important reason is that many features extracted by the computer are not visible by human vision or the meanings represented by these features are inaccessible to human understanding. It is almost impossible to directly interpret the...

Database Selection

Since the topology of an ANN or a BBN depends on a set of features and the distribution of weights inside the network depends on a set of training data, both the database and selected features play important roles in determining the training and testing performance of the network. Figure 3 shows a schematic representation of the process of evaluating the performance of a classifier using either an ANN or a BBN in medical image processing. In the figure, collection of sample data (or images) is the first step for building and testing a machine learning classification system. In this section, several issues related to the database selection are discussed.


The stability of the tag pattern within the myocardium leads to a 2D apparent motion of the tag pattern within the image plane, despite the fact that the true motion is 3D. The concept of apparent motion proves very useful for describing tagged image processing. To define apparent motion mathematically, we first define a 2D reference position given by

Pixel Operations

In this section we present methods of image enhancement that depend only upon the pixel gray level and do not take into account the pixel neighborhood or whole-image characteristics. Intensity scaling is a method of image enhancement that can be used when the dynamic range of the acquired image data significantly exceeds the characteristics of the display system, or


FIGURE 3 (a) The convergence of a deformable contour using (b) GVF external forces, (c) shown close-up within the boundary concavity. Reprinted from C. Xu and J. L. Prince, Snakes, shapes, and gradient vector flow. IEEE Trans, on Image Processing, 7(3) 359-369, March, 1998. 1998 IEEE. FIGURE 3 (a) The convergence of a deformable contour using (b) GVF external forces, (c) shown close-up within the boundary concavity. Reprinted from C. Xu and J. L. Prince, Snakes, shapes, and gradient vector flow. IEEE Trans, on Image Processing, 7(3) 359-369, March, 1998. 1998 IEEE.

Concluding Remarks

After these algorithms have been applied and adjusted for best outcome, additional image enhancement may be required to improve image quality further. Computationally more intensive algorithms may then be considered to take advantage of context-based and object-based information in the image. Examples and discussions of such techniques are presented in subsequent chapters.

[1 Arrays of Transfected Mammalian Cells for High Content Screening Microscopy

Recent advances in automated fluorescence scanning microscopy and image processing (see e.g., Liebel et al., 2003 Starkuviene et al., 2004) allow now rapid analysis of transfected cell arrays in large scale screening applications. In the following we describe the method of reverse transfection on cell arrays as we use it in our laboratory to examine gene function by RNAi or overexpression of plasmid DNAs with high content screening microscopy.

Minimal Path Approach Model

Another drawback with MPA is that this approach lacks the topography handling ability. For some applications within our study, such as carotid artery lumen contour tracking in MRI sequences, the topology of blood vessel lumen in each cross-section images may change due to bifurcation, and it is impossible to apply MPA directly even though the initial points can be provided precisely. Therefore, a mechanism is needed to track the topology changes for automatic image processing.

Segmentation of Contrast Enhanced Digitized Mammograms

The aim of image segmentation is to label a pixel in an image as belonging to one of the known corresponding real world objects. In the detection of breast lesions in digitized mammograms, image segmentation results in contiguous areas or regions of pixels, labeled as normal or suspicious. For the purpose of evaluating image enhancement, we use an unsupervised Gaussian mixture model (GMM) and hidden Markov random field (HMRFu) model of image segmentation proposed by Zhang et al. 24 . For ease of referencing, this shall be referred to as HMRFU in the rest of this chapter. The HMRFU segmentation method is used to segment contrast-enhanced images so that the performance of the contrast enhancement can be determined. The HMRFU segmentation algorithm operates in an unsupervised manner. The only a priori knowledge required for the segmentation is the maximum number of classes, L, from which a pixel is labelled. By setting L 2, HMRFU will label pixels as...

Helical CT Imaging Characteristics

Mri Cross Section Anatomy Elbow

Important to the development of image processing techniques for pancreatic cancer applications is the knowledge of the clinical imaging characteristics of the normal and abnormal pancreas. The normal pancreas is relatively easy to delineate on CT slices. Understanding how the image of the normal pancreas may be distorted by disease and particularly pancreatic masses (benign or malignant) is the basis for selecting robust features for the development of automated segmentation, classification, registration, and reconstruction methodologies. The most important features used by the radiologists and oncologists in the evaluation of pancreatic adenocarcinoma on radiologic images are summarized in Table 4.1. These features are merely general observation that may not always hold.

Knowledge Representation by Image Grouping on Various Criteria

Matsubara et al. 10 proposed the use of an image grouping scheme for digitzed mammograms. In their study, images are assigned to one of four categories based on histogram analysis of the image gray scales. Subsequent image-processing operations, such as threshold-based segmentation and region classification operate on parameters defined empirically and independently within each category. The authors use this scheme to ignore high-density mammograms. On a small dataset of 30 images, the authors report a sensitivity of 93 .

Bone Density Assessment

Histologic analysis starts with fixed tissues that are then decalcified and stained for H&E or prepared and histomorphometrically analyzed. Growth plate measurements can be made on tibial samples stained with alcian blue Van Gieson stain and sectioned to 4 um thick. An image-processing system is used coupled to a microscope. Histologic analysis can help distinguish osteoporosis from osteomalacia. Osteomalacia is a failure to mineralize versus osteoporosis which is a reduction in bone mass. Bone histomorphometry is measured as the ratio of trabecular bone volume to total volume. Areas of trabecular bone within a reference area are stained with H&E and measured in sections. Measurements are made on printed copies by point counting with a square lattice.38 Enzyme-histochemical staining can be done by staining for alkaline phosphatase activity or tartrate-resistant acid phosphatase. Serum assays are used to determine total protein, calcium, phosphorus, and creatinine. Serum osteocalcin, a...

Intraoperative Video Display

Once the images have been registered with respect to the patient, they must be displayed in a meaningful manner. A high-resolution monitor with 512 X 512 pixel windows displays images in a variety of orientations and configurations. Color graphic overlays that represent the localizer position and trajectory are usually displayed relative to the on-screen images. The local-izer position is usually updated on the screen at 20 frames per second. Realtime displays have become more common with increases in image processing speed and power and decreases in cost.

Imaging And Targeting

Gantry tilt should be avoided if a surgical navigation (SN) computer or the Radionics Stereo Calc program is to be used for image processing if the dedicated Radionics mini-computer (MC) is used, then the gantry may be angled to optimize target visualization. This will be of use mainly for identifying the AC-PC line for functional targeting.

Computer Assisted Techniques for Skeletal Determinations

With the advent of digital imaging, several investigators have attempted to provide an objective computer-assisted measure for bone age determinations and have developed image processing techniques from reference databases of normal children that automatically extract key features of hand radiographs 13-17 . To date, however, attempts to develop automated image analysis techniques capable of extracting quantitative measures of the morphological traits depicting skeletal maturity have been hindered by the inability to account for the great variability in development and ossification of the multiple bones in the hand and wrist. In an attempt to overcome these difficulties, automated techniques are being developed that primarily rely on measures of a few ossification centers, such as those of the epiphy-ses. Our aim was to provide a portable alternative to the reference books currently available, while avoiding the complexity of computer assisted image analysis. The wide adoption of...

Structural studies on gastroenteritis viruses

There are many recent advances in our understanding of the structure-function relationships in rotavirus, a major pathogen of infantile gastroenteritis, and Norwalk virus, a causative agent of epidemic gastroenteritis in humans. Rotavirus is a large (1000 A) and complex icosahedral assembly formed by three concentric capsid layers that enclose the viral genome of 11 dsRNA segments. Because of its medical relevance, intriguing structural complexity, and several unique strategies in the morphogenesis and replication, this virus has been the subject of extensive biochemical, genetic and structural studies. Using a combination of electron cryomicroscopy and computer image processing together with atomic resolution X-ray structural information, we have been able to provide not only a better description of the rotavirus architecture, but also a better understanding of the structural basis of various biological functions such as trypsin-enhanced infectivity, virus assembly and the...

Cartesian Genetic Programming

CGP has been applied to a growing number of domains and problems digital circuit design (Miller et al., 2000a, Miller et al., 2000b), digital filter design (Miller, 1999), image processing (Sekanina, 2004), artificial life (Rothermich and Miller, 2002), bio-inspired developmental models (Miller and Thomson, 2003, Miller, 2003, Miller and Banzhaf, 2003), evolutionary art (Ashmore, 2000) and has been adopted within new evolutionary techniques cell-based Optimization (Rothermich et al., 2003) and Social Programming (Voss, 2003, Voss and James C. Howland, 2003).

Image Contrast Enhancement Layer

In order to construct a scheme for the optimal selection of image enhancement, some quantitative indices are needed that measure the amount of enhancement. Not enough research has been conducted to tackle this difficult issue. In our previous work 19, 20 we introduced three new quantitative measures of image enhancement based on the change in contrast between the target (mass) and the backgound (a border 20 pixels wide around the target). We cover these measures for the sake of completeness here in section 11.3.1. In addition, we also discuss

Evaluation of the Knowledge Based Model

This section evaluates the performance of a given configuration of the adaptive knowledge-based model in predicting the optimal pipeline of image processing operators used for the CAD of breast cancer. This performance is compared to that obtained by keeping the pipeline fixed. Contrast enhancement and image segmentation are the key components in a mammographic CAD system. For these key components, sections 11.3 and 11.4, respectively, have demonstrated that a knowledge-based framework is superior to the single best method in each case. Parameterized versions of these components have been engineered for individual mammogram groupings. These groupings are based on the mammographic breast density and a mechanism for its prediction. Evaluation of the performance of each parameterized version of the knowledge-based component presented in the previous sections has been performed using the target mammogram breast grouping. In this section, the complete adaptive knowledge-based model is...

Contrast Enhancement Mixture of Experts Framework

This section describes the mixture of experts framework and it is laid out as follows. Section reviews the contrast enhancement experts used to build the framework. Then the segmentation algorithm used to evaluate the enhanced images is briefly described together with quantitative measures of segmentation performance. In section results are presented when applying the different image enhancement on DDSM images and the resulting segmentation from them. Section discusses the features that can be extracted from the mammograms to be fed into a mapping scheme (e.g., neural networks) that maps features to optimal enhancement methods. Finally, section discusses a machine learning system for this mapping. A neural network is used in two different modes double network mapping and a single direct mapping scheme.

Application of Image Analysis to the Diagnosis of Diabetic Retinopathy

Image enhancement Images taken at standard examinations are often noisy and poorly contrasted. Over and above that illumination is normally not uniform. Techniques improving contrast and sharpness and reducing noise are therefore required. 2. Mass screening Computer-assisted mass screening is certainly the most important task to which image processing can contribute. We have already seen that the blinding complication of diabetic retinopathy can be inhibited by early treatment. However, as vision normally alters only in the later stages of the disease, many patients remain undiagnosed in the earlier stages of the disease. Hence, mass screening of all diabetic patients would help to diagnose this disease early enough. Unfortunately, this approach is not very realistic, taking into consideration the large number of diabetic patients compared to a lack to specialists. Computer assistance could make mass screening a lot more efficient. Of course, giving detailed solutions to all these...

Computer Applications in Pancreatic Cancer Imaging

There is limited development of automatic approaches for the detection and or diagnosis of pancreatic cancer either from CT or other imaging modalities. This is certainly an area worthy of further investigation and an area identified as in great need of technological advances by the NCI Review Group 2 . Imaging priorities set by the Group have been summarized earlier in this chapter. One of the most interesting recommendation was for a collaborative research and training approach that will link molecular biology, pathology, and imaging as well as for a well documented source of images to support computer applications and image processing 2 . Figure 4.5 General algorithm design for CT image processing. Processing may include a segmentation, a classification, a registration, a reconstruction step, or any combinations of these. Figure 4.5 General algorithm design for CT image processing. Processing may include a segmentation, a classification, a registration, a reconstruction step, or...

An Adaptive Knowledge Based Model

As we can see from the survey in the previous section, little research has been undertaken in attempting to optimize a hierarchy of image processing operators. In this chapter an adaptive knowledge-based model is proposed. The model comprises deterministic and knowledge-based components for the detection of

Nextgeneration Experimental Systems

In the developmental biology of Caenorhabditis elegans, identification of cell lineage is one of the major issues that needs to be met to assist analysis of the gene regulatory network for differentiation. The first attempt to identify cell lineage was carried out entirely manually 31, 32 and required several years to identify the lineage of the wild type. Four-dimensional microscopy allows multilayer confocal images to be collected at constant time intervals, but lineage identification is not automatic. With the availability of exhaustive RNAi knockout C. elegans, high-throughput cell lineage identification is essential for exploring the utility of the exhaustive RNAi. Efforts are under way to fully automate cell lineage identification, as well as acquisition of three-dimensional nuclear position data 33 , fully utilizing advanced image-processing algorithms and massively parallel supercomputers. Such devices meet some ofthe criteria mentioned earlier and provide comprehensive...

Methods and Techniques

For each age group, prior to creating a composite idealized image from the different selected key images, three image processing steps and enhancements were applied for standardization. First, the background was replaced by a uniform black setting and the image size was adjusted to fit into square images of 800 x 800 pixels. Second, contrast and intensity were optimized using predefined window and level thresholds. Lastly, the image was processed through a special edge enhancement filter based on an unsharp masking algorithm tailored to provide optimum sharpness of bone structure for hand-held devices.

Present Limitations of the IVUS Technique and the Need for a Generation Model of IVUS Data

The mentioned shortcomings are difficult to quantify and depend on the experience of the operator, that is he should have been trained in handling a large number of patient cases. Some of the limitations of the IVUS technique can be attenuated through algorithms of image processing the limitations due to a suboptimal location of the borders of the arterial structure can be overcome with new algorithms of segmentation. The question is how to develop robust algorithms that can solve these problems, analyzing the artifacts with their multiple appearances in IVUS images. Having a complete set of patient data to present all variance of artifacts appearance in images would mean to dispose of a huge number of patient cases. A more efficient solution is to develop a simulation model for IVUS data construction so that synthetic data is available in order to train image processing techniques. In this way, different appearances of artifacts can be designed to assure robust performance of image...

Semiimplicit CoVolume Scheme

First we choose a uniform discrete time step t and a variance a of the smoothing kernel Ga. Then we replace time derivative in (11.8) by backward difference. The nonlinear terms of the equation are treated from the previous time step while the linear ones are considered on the current time level, this means semi-implicitness of the time discretization. In the last decade, semi-implicit schemes have become a powerful tool in image processing, we refer e.g. to 3,4,25-27,33,37,51,57,58 .

Image Registration Using Wavelets

In this section, we give a brief overview of another very important application of wavelets in image processing image registration. Readers interested in this topic are encouraged to read the references listed in the context. Image registration is required for many image processing applications. In medical imaging, co-registration problems are important for many clinical tasks

Phase Contrast PC Technique

4D PC technique was demonstrated for its feasibility that permits spatial and temporal coverage of an entire 3D volume 26 . It validated quantitatively the accuracy against an established time resolved 2D PC technique to explore advantages of the approach with regard to the 4D nature of the data. Time-resolved, 3D anatomical images were generated simultaneously with registered three-directional velocity vector fields. Improvements were compared to prior methods for gated and respiratory compensated image acquisition, interleaved flow encoding with freely selectable velocity encoding (VENC) along each spatial direction, and flexible trade-off between temporal resolution and total acquisition time. The implementation was validated against established 2D PC techniques using a well-defined phantom, and successfully applied in volunteer and patient examinations. Human studies were performed after contrast administration in order to compensate for loss of in-flow enhancement in the 4D...

Survey of Plaque Segmentation Techniques

Figure 9.3 shows the different image processing techniques used for segmentation of the plaque volumes. Yuan et al. 33 used a quantitative vascular analysis tool (QVAT). The QVAT is a semiautomatic, custom-designed program that tracks boundaries and computes areas. Gill et al. 9 used a mesh-based model that obtained boundaries in three steps. It involved a deformable balloon model of a triangular mesh which is first placed inside a region manually it is then inflated by inflation forces and then refined by image-based forces. Kim et al. 11 used an edge-detection tool. Wilhjelm et al. 12 used a manual segmentation procedure. Yang et al. 10 used a border-based model, which had three steps. It involved first approximating the outlines of the vessels, followed by the detection of borders, and then the user correction of the borders.

Level Set Surface Deformation

Edges Conventional edge detectors from the image processing literature produce sets of edge voxels that are associated with areas of high contrast. For this work we use a gradient magnitude threshold combined with nonmaximal suppression, which is a 3D generalization of the method of Canny 16 . The edge operator typically requires a scale parameter and a gradient threshold. For the scale, we use small, Gaussian kernels with standard deviation a 0.5,1.0 voxel units. The threshold depends on the contrast of the volume. The distance transform on this edge map produces a volume that has minima at those edges. The gradient of this volume produces a field that attracts the model to these edges. The edges are limited to voxel resolution because of the mechanism by which they are detected. Although this fitting is not sub-voxel accurate, it has the advantage that it can pull models toward edges from significant distances, and thus inaccurate initial estimates can be brought into close...

Wavelet Transform and Multiscale Analysis

One of the most fundamental problems in signal processing is to find a suitable representation of the data that will facilitate an analysis procedure. One way to achieve this goal is to use transformation, or decomposition of the signal on a set of basis functions prior to processing in the transform domain. Transform theory has played a key role in image processing for a number of years, and it continues to be a topic of interest in theoretical as well as applied work in this field. Image transforms are used widely in many image processing fields, including image enhancement, restoration, encoding, and description 12 .

Inflow Method Time of Flight

This method belongs to a class of MR angiographic techniques known as time-of-flight. This technique gives rise to 3D information about the vessels in the volume of tissue being imaged with high contrast between the stationary tissue and the flowing blood. The INFLOW method relies on the flow related image enhancement caused by the movement of fresh, unsaturated blood into an already saturated slab of tissue. The INFLOW method has a number of advantages over other angiographic imaging methods. First, image subtraction is not necessary, thereby reducing scan time and computing requirements while speeding data manipulation. Second, high contrast can be obtained virtually independent of flow velocity. Third, the arteries or veins may be selectively imaged by the use of presaturation slabs. Finally, the technique does not require the use of self-shielded gradients. It is less sensitive to motion than the phase contrast methods. Using the INFLOW technique, angiograms may be obtained in...

F Developing and printing

Edited, cropped or colored using the latest versions of Adobe Photoshop or Paint Shop Pro and presented on Microsoft Power Point. The latter is useful for labeling and annotating for presentations. Most images are saved on Tiff format and later converted to JPEG or GIF for online publication or transmission by email or the web (see for some images).

Detection Segmentation Stage

The terms segmentation and detection may be confusing for the reader not so familiar with the medical imaging vernacular. In some instances these terms may be used interchangeably, but other times not. We might consider segmentation as being a more refined or specialized type of detection. For instance, we may gate a receiver for some time increment and make a decision as to whether or not a signal of interest was present within the total time duration, but not care about exactly where the signal is within the time window this may be defined as a detection task with a binary output of yes or no. Segmentation takes this a step further. With respect to the image processing, the detection task makes a decision if the abnormality is present, which in this case is a calcification. If, in addition, the detection provides some reasonable estimate as to the spatial location and extent of the abnormality, then we would say that the calcification

Conclusion and Perspectives

In this chapter, we have seen different ways of computer assistance to the diagnosis of diabetic retinopathy, which is a very frequent and severe eye-disease image enhancement, mass screening, and monitoring. Different algorithms within this framework have been presented and evaluated with encouraging results.

Wavelet Based Methods

Wavelet theory has been enthusiastically adopted by people in the area of signal and image processing. It has been proved to be a useful tool in many applications. A wavelet-based shape from shading method was introduced in 31 . Unlike methods introduced in Section 5.3, the objective function in the constrained optimization problem is replaced by its projection to the wavelet subspaces. To understand this approach, we first recall some elements in wavelet theory.

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 filtering. The only assumption we make here is that the reader understands Fourier analysis in one dimension. If so, the extension to two dimensions will be easily accomplished.

MRI and Atlas Fusion

Image processing software is required during the preplanning process to fuse the CT and MRI scans. The ideal software program calibrates an individual patient's imaging studies to three standard neurosurgical atlases, the Talairach, Schaltenbrand, and Watkins, and warps the atlases until they reflect the patient's anatomy (Fig. 1). There are nomograms to correct for individual variations in anatomy. This set of images can then be plugged into any or all of several surgical planning modules that provide for comprehensive planning of trajectories, designed to optimize the intervention and minimize damage to uninvolved tissue.

The Editors

He has chaired image processing tracks at several international conferences and has given more than 40 international presentations seminars. Dr. Suri has written four books in the area of body imaging (such as cardiology, neurology, pathology, mammography, angiography, atherosclerosis imaging) covering medical image segmentation, image and volume registration, and physics of medical imaging modalities like MRI, CT, X-ray, PET, and ultrasound. He also holds several United States patents. Dr. Suri has been listed in Who's Who seven times, is a recipient of president's gold medal in 1980, and has received more than 50 scholarly and extracurricular awards during his career. He is also a Fellow of American Institute of Medical and Biological Engineering (AIMBE) and ABI. Dr. Suri's major interests are computer vision, graphics and image processing (CVGIP), object oriented programming, image guided surgery and teleimaging. Dr. Suri had worked for Philips Medical Systems and Siemens Medical...


From image processing point of view, segmentation, the process of grouping image pixels into a collection of subregions or partitions that are statistically homogeneous with respect to one or more characteristics, such as intensity, color, texture, etc., has been a very important region analysis technique in medical image applications. The eventual goal of segmentation is to aggregate those neighboring pixels with similar features as a region and separate it from the others or the background in the image. Since the partitioned regions sometimes do not contain any semantic meaning corresponding to the real physical object in image, image segmentation technique often serves as a low-level processing step in image-processing procedures. However, it is very crucial to the success of higherlevel recognition process and plays as a deterministic role to the eventual performance.

Markov Random Field

MRF has become a significant statistical signal modeling technique in image processing and computer vision. Generally speaking, the MRF model assumes that the information contained in a particular location is affected by its neighboring local structure of a given image rather than the whole image. In other words, the estimation of pixel's properties, such as intensity, texture, color, etc., closely relates to a neighborhood of pixels, and this dependency can be characterized by means of a local conditional probability distribution. This hypothesis can

Computed Tomography

An important advantage of helical CT is the ability to reconstruct the image data acquired on the initial scan at intervals as small as 1 mm. Such reconstruction can improve lesion conspicuity by placing the lesion directly within the image plane rather than volume-averaging it between 2 contiguous reconstructed images. Smaller lesions can therefore be detected with helical CT. In addition, the reconstructed images can be stacked to form a volume of image data so that they can be displayed in multiple planes or in a 3-dimensional format. This image processing technique forms the basis for CT colonography, CT angiography, and CT cholangiography.

Chapter Overview

Diagnostic imaging studies provide important information in the diagnostic and staging evaluation of patients with gastrointestinal malignancies. Advances in imaging techniques, contrast development, and image-processing techniques continue to improve our ability to display images, make a correct diagnosis, and accurately stage disease. To maximize the potential of advanced imaging techniques, imaging protocols must be designed properly. Proper interpretation is also critical to maximizing the clinical benefit of imaging studies.


A 2D basis constructed from the tensor product of 1D wavelet basis is much easier to compute than the nonseparable wavelets. There is also some ongoing research on nonseparable wavelets for use in image processing. For a detailed discussion on nonseparable wavelets, we recommend 37,38,40 and references therein.

Further Reading

Deformable contour models are commonly used in image processing and computer vision, for example for shape description 21 , object localization 22 , and visual tracking 23 . Level set methods can be computationally expensive. A number of fast implementations for geometric snakes have been proposed. The narrow band technique, initially proposed by Chop 31 , only deals with pixels that are close to the evolving zero level set to save computation. Later, Adalsterinsson et al. 32 analyzed and optimized this approach. Sethian 33,34 also proposed the fast marching method to reduce the computations, but it requires the contours to monotonically shrink or expand. Some effort has been expended in combining these two methods. In 35 , Paragios et al. showed this combination could be efficient in application to motion tracking. Adaptive mesh techniques 36 can also be used to speed up the convergence of PDEs. More recently, additive operative splitting (AOS) schemes were introduced by Weickert et...

The Algorithm

In multichannel data image processing, different channels usually convey different amount of information. For example, in the soft tissue type identification with MR imaging 81 , subjects are generally scanned with the multiple contrast weightings, such as T1-weighted (T1W), T2-weighted (T2W), proton density-weighted (PDW), and 3D time-of-flight (3D TOF). Since each contrast weighting imaging technique is sensitive only to certain tissue types, therefore, they usually contribute differently to the final decision when different tissue type is analyzed.

E Microscopy

Thick sections are examined with a light microscope, while advanced imaging is done with a research microscope with a digital camera, hooked to a computer for image processing and editing. We use Leica QWin, Nikon or Olympus digital microscopes. Alternatively, the sections can be photographed on film and printed or mounted on 35 mm slides. The resolution of photographic images is superior to that of the digital images and they could be scanned on to a computer for editing. Thin sections are examined by TEM. We use Jeol, Hitachi or Philips microscopes but others such as Zeiss are equally good. The transmission electron microscope has to be maintained, and operated by an experienced technician. Lower magnifications (x 2,000 to x 5,000) are more useful to image whole cells, while

Image Segmentation

Some of these types of images can be pseudocolored. As mentioned earlier in section 6.1, most of the segmentation techniques developed for gray scale images can be extended to color images. There are quite a few color models 17 that are commonly used in image processing, mainly to comply with color video standards and human perception. RGB (red, green, blue), HSI (hue, saturation, intensity), and CIE L*a*b* are color models that have been frequently used in segmentation. RGB is hardware oriented while HSI and L* a*b* representations are compatible with human visual perception. What is more, perceptual uniformity of the L*a*b* color space is advantageous over RGB and HSI in that the human perception of color difference can be represented as the euclidean distance between color points, a useful property can be used in error functions of some segmentation algorithms. Most color images, such as the ones used in our examples, the color retinal stereo images and the color...


In this chapter we discuss a basic physical model to generate synthetic 2D IVUS images. The model has different utilities Firstly, an expert can generate simulated IVUS images in order to observe different arterial structures of clinical interest and their gray-level distribution in real images. Secondly, researchers and doctors can use our model to learn and to compare the influence of different physical parameters in the IVUS image formation, such as the ultrasound frequency, the attenuation coefficient, the beam number influence, and the artifact generations. Thirdly, this model can generate a large database of synthetic data under different device and acquisition parameters to be used for validating the robustness of image processing techniques. The IVUS image generation model provides a basic methodology that allows us to observe the most important real image emulation aspects. This initial phase does not compare pixel to pixel values generation, showing the coincidence with the...

Learn Photoshop Now

Learn Photoshop Now

This first volume will guide you through the basics of Photoshop. Well start at the beginning and slowly be working our way through to the more advanced stuff but dont worry its all aimed at the total newbie.

Get My Free Ebook