4-D Visualization

ABSTRACT:

     Visualizable objects in biology and medicine extend across a vast range of scale, from individual molecules and cells through the varieties of tissue and interstitial interfaces to complete organs, organ systems, and body parts. The practice of medicine and study of biology have always relied on visualizations to study the relationship of anatomic structure to biologic function and to detect and treat disease and trauma that disturb or threaten normal life processes. Traditionally, these visualizations have been either direct, via surgery or biopsy, or indirect, requiring extensive mental reconstruction. The potential for revolutionary innovation in the practice of medicine and in biologic investigations lies in direct, fully immersive, real-time multi sensory fusion of real and virtual information data streams into online, real-time visualizations available during actual clinical procedures or biological experiments. In the field of scientific visualization, the term "four dimensional visualization" usually refers to the process of rendering a three dimensional field of scalar values. "4D" is shorthand for "four-dimensional"- the fourth dimension being time. 4D visualization takes three-dimensional images and adds the element of time to the process. The revolutionary capabilities of new three-dimensional (3-D) and four-dimensional (4-D) medical imaging modalities along with computer reconstruction and rendering of multidimensional medical and histologic volume image data, obviate the need for physical dissection or abstract assembly of anatomy and provide powerful new opportunities for medical diagnosis and treatment, as well as  for biological investigations.In contrast to 3D imaging diagnostic processes, 4D allows doctor to visualize internal anatomy moving in real-time. So physicians and sonographers can detect or rule out any number of issues, from vascular anomalies and genetic syndromes. Time will reveal the importance of 4d visualization.

                                              INDEX:

SR NO	CONTENT	PAGE NO.
1	INTRODUCTION	1
2	3D- IMAGE GENERATION, DISPLAY AND VISUALIZATION	3
3	CONCEPT OF 4D VISUALIZATION	6
4	VOLOCITY- A RENDERING SYSTEM	9
5	4D VISUALIZATION IN LIVING CELLS	12
6	WORKSTATION FOR ACQUISITION, RECONSTRUCTION AND VISUALIZATION OF 4D IMAGES OF HEART.	15
7	4D IMAGE WARPING FOR MEASUREMENT OF LONGITUDINAL BRAIN CHANGES	17
8	MED-SANAREA: BRIGHT FUTURE	18
9	BIBLIOGRAPHY	19

INTRODUCTION:

     The practice of medicine and study of biology have always relied on visualizations to study the relationship of anatomic structure to biologic function and to detect and treat disease and trauma that disturb or threaten normal life processes. Traditionally, these visualizations have been either direct, via surgery or biopsy, or indirect, requiring extensive mental reconstruction. The revolutionary capabilities of new three-dimensional (3-D) and four-dimensional (4-D) medical imaging modalities [computed tomography (CT), magnetic resonance imaging (MRI), positron emission tomography (PET), ultrasound (US), etc. along with computer reconstruction and rendering of multidimensional medical and histologic volume image data, obviate the need for physical dissection or abstract assembly of anatomy and provide powerful new opportunities for medical diagnosis and treatment, as well as for biological investigations.

4D-THE MODERN DIMENSION:

    "4D" is shorthand for "four-dimensional"- the fourth dimension being time. 4D visualization takes three-dimensional images and adds the element of time to the process.
In contrast to 3D imaging diagnostic processes, 4D allows doctor to visualize internal anatomy moving in real-time. For example: Movement patterns of fetuses allows conclusions to be drawn about their development; increase of accuracy in ultrasound guided biopsies thanks to the visualization of needle movements in real time in all 3 planes. So physicians and sonographers can detect or rule out any number of issues, from vascular anomalies and genetic syndromes.

 
3D GIVES LIFE TO 4D:

     Locked within 3-D biomedical images is significant information about the objects and their properties from which the images are derived. Efforts to unlock this information to reveal answers to the mysteries of form and function are couched in the domain of image processing and visualization. A variety of both standard and sophisticated methods have been developed to process (modify) images to selectively enhance the visibility and measurability of desired object features and properties. For example, both realism-preserving and perception-modulating approaches to image display have significantly advanced the practical usefulness of  4-D biomedical imaging.

     Many life-threatening diseases and/or quality-of-life afflictions still require physical interventions into the body to reduce or remove disease or to alleviate harmful or painful conditions. But minimally invasive or noninvasive interventions are now within reach that effectively increase physician performance in arresting or curing disease; reduce risk, pain, complications, and reoccurrence for the patient; and decrease healthcare costs. What is yet required is focused reduction of recent and continuing advances in visualization technology to the level of practice, so that they can provide new tools and procedures that physicians ‘‘must have’’ to treat their patients and empower scientists in biomedical studies of structure-to function relationships.

Forming an image is mapping some property of an object onto image space. This space is used to visualize the object and its properties and may be used to characterize quantitatively its structure or function. Imaging science may be defined as the study of these mappings and the development of ways to better understand them, to improve them, and to use them productively. The challenge of imaging science is to provide advanced capabilities for acquisition, processing, visualization, and quantitative analysis of biomedical images to increase substantially the faithful extraction of useful information that they contain.

The particular challenge of imaging science in biomedical applications is to provide realistic and faithful displays, interactive manipulation and simulation, and accurate, reproducible measurements.  The goal of visualization in biomedical computing is to formulate and realize a rational basis and efficient architecture for productive use of biomedical-image data. The need for new approaches to image visualization and analysis will become increasingly important and pressing as improvements in technology enable more image data of complex objects and processes to be acquired. The value of such visualization technology in medicine will derive more from the enhancement of real experience than from the simulation of reality. Visualizable objects in medicine extend across a vast range of scale, from individual molecules and cells, through the varieties of tissue and interstitial interfaces, to complete organs, organ systems, and body parts, and these objects include functional attributes of these systems, such as biophysical, biomechanical, and physiological properties. Medical applications include accurate anatomy and function mapping, enhanced diagnosis, and accurate treatment planning and rehearsal. However, the greatest potential for revolutionary innovation in the practice of medicine lies in direct, fully immersive, real-time multisensory fusion of real and virtual information data streams into an online, real-time visualization during an actual clinical procedure. Such capabilities are not yet available to the general practitioner. However, current advanced computer image-processing research has recently facilitated major progress toward fully interactive 3-D visualization and realistic simulation. The continuing goals for development and acceptance of important visualization display technology are (a) improvement in speed, quality, and dimensionality of the display

and (b) improved access to the data represented in the display through interactive,intuitive manipulation and measurement of the data represented by the display.

Included in these objectives is determination of the quantitative information about the properties of anatomic tissues and their functions that relate to and are affected by disease. With these advances in hand, the delivery of several important clinical applications will soon be possible that will have a significant impact on medicine and study of biology.

3D- IMAGE GENERATION, DISPLAY AND VISUALIZATION

Display and visualization are not fully synonymous. Visualization of 3-D biomedical-volume images has traditionally been divided into two different techniques:

1. surface rendering
2. volume rendering.

Both techniques produce a visualization of selected structures in the 3-D–volume image, but the methods involved in these techniques are quite different, and each has its advantages and disadvantages. Selection between these two approaches is often predicated on the particular nature of the biomedical-image data, the application to which the visualization is being applied, and the desired result of the visualization.

Surface Rendering:

 Surface-rendering techniques characteristically require the extraction of contours (edges) that define the surface of the structure to be visualized. An algorithm is then applied that places surface patches or tiles at each contour point, and, with hidden surface removal and shading, the surface is rendered visible.

ADVANTAGE:

v  Relatively small amount of contour data, resulting in fast rendering speeds.

v  Standard computer graphics techniques can be applied, including shading models (Phong, Gouraud).

v  The contour-based surface descriptions can be transformed into analytical descriptions, which permit use

v  with other geometric-visualization packages [e.g., computer-assisted design and manufacturing (CAD/CAM) software]

v  Contours can be used to drive machinery to create models of the structure.

v  Other analytically defined structures can be easily superposed with the surface-rendered structures.

DISADVANTAGES:

v  Need to discretely extract the contours defining the structure to be visualized.

v  Other volume image information is lost in this process, which may be important for slice generation or value measurement.

v  Any interactive, dynamic determination of the surface to be rendered is prohibited, because the decision has been made during contour extraction regarding specifically which surface will be visualized.

v  Due to the discrete nature of the surface patch placement, surface rendering is prone to sampling

v  and aliasing artifacts on the rendered surface.

.

Volume Rendering:

     One of the most versatile and powerful image display and manipulation techniques is volume rendering. Volume-rendering techniques based on ray-casting algorithms have generally become the method of choice for visualization of 3-D biomedical volume images.

     Ray-tracing model is used to define the geometry of the rays cast through the scene(volume of data). To connect the source point to the scene, for each pixel of the screen, a ray is defined as a straight line from the source point passing through the pixel. To generate the picture, the pixel values are assigned appropriate intensities ‘‘sampled’’ by the rays passing everywhere through the scene. For instance, for shaded surface display, the pixel values are computed based on light models (intensity and orientation of light source(s), reflections, textures, surface orientations, etc.) where the rays have intersected the scene.

     There are two general classes of volume display: transmission and reflection.  For transmission-oriented displays, there is no surface identification involved. A ray passes totally through the volume, and the pixel value is computed as an integrated function. There are three important display subtypes in this family: brightest voxel, weighted summation, and surface projection (projection of a thick surface layer). For all reflection display types, voxel density values are used to specify surfaces within the volume image. Three types of functions may be specified to compute the shading—depth shading, depth gradient shading, and real gradient shading.

     Full-gradient volume-rendering methods can incorporate transparency to show two different structures in the display, one through another one. The basic principal is to define two structures with two segmentation functions. To accomplish this, a double threshold on the voxel density values is used. The opaque and transparent structures are specified by the thresholds used. A transparency coefficient is also specified. The transparent effect for each pixel on the screen is computed based on a weighted function of the reflection caused by the transparent structure, the light transmission through that structure, and the reflection of the opaque structure.

    

Examples of interactive volume-rendering operations, including selective surfaces, cut-planes, orthogonal dissections, and render masks, which permit mixing of rendering types (e.g., a transmission projection within a reflection surface.)

ADVANTAGES:

v  Direct visualization of the volume images without the need for prior surface or object segmentation, preserving the values and context of the original image data.

v  Application of various different rendering algorithms during the ray-casting process.

v  Surface extraction is not necessary, as the entire volume image is used in this rendering process, maintaining the original volume image data.

v  Capability to section the rendered image and visualize the actual image data in the volume image and to make voxel-value based measurements for the rendered image.

v  The rendered surface can be dynamically determined by changing the ray-casting and surface recognition conditions during the rendering process.

v  Display surfaces with shading and other parts of the volume simultaneously.

v  Displays data directly from the gray-scale volume.

.

CONCEPT OF 4D VISUALIZATION:

     In the field of scientific visualization, the term "four dimensional visualization" usually refers to the process of rendering a three dimensional field of scalar values. While this paradigm applies to many different data sets, there are also uses for visualizing data that correspond to actual four-dimensional structures. Four dimensional structures have typically been visualized via wire frame methods, but this process alone is usually insufficient for an intuitive understanding. The visualization of four dimensional objects  is possible through wire frame methods with extended visualization cues, and through ray tracing methods. Both the methods employ true four-space viewing parameters and geometry. The ray tracing approach easily solves the hidden surface and shadowing problems of 4D objects, and yields an image in the form of a three-dimensional field of RGB values, which can be rendered with a variety of existing methods. The 4D ray tracer also supports true four-dimensional lighting, reflections and refractions.

     The display of four-dimensional data is usually accomplished by assigning three dimensions to location in three-space, and the remaining dimension to some scalar property at each three-dimensional location. This assignment is quite apt for a variety of four-dimensional data, such as tissue density in a region of a human body, pressure values in a volume of air, or temperature distribution throughout a mechanical object.

Viewing in Three-Space

     The first thing to establish is the viewpoint, or viewer location. This is easily done by specifying a 3D point in space that marks the location of the viewpoint. This is called the from-point or viewpoint.

     3D Viewing Vectors and From, To Points            The Resulting View  

     The next thing to establish is the line of sight. This can be done by either specifying a line-of-sight vector, or by specifying a point of interest in the scene. The point-of-interest method has several advantages. One advantage is that the person doing the rendering usually has something in mind to look at, rather than some particular direction. It also has the advantage that you can ``tie'' this point to a moving object, so we can easily track the object as it moves through space. This point of interest is called the to-point.Now to pin down the orientation of the viewer/scene ,a vector is specified that will point straight up after being projected to the viewing plane. This vector is called the up-vector.

     Since the up-vector specifies the orientation of the viewer about the line-of-sight, the up-vector must not be parallel to the line of sight. The viewing program uses the up-vector to generate a vector orthogonal to the line of sight and that lies in the plane of the line of sight and the original up-vector. 

                                                                                         

If we're going to use perspective projection, we need to specify the amount of perspective, or ``zoom'', that the resultant image will have. This is done by specifying the angle of the viewing cone, also known as the viewing frustum. The viewing frustum is a rectangular cone in three-space that has the from-point as its tip, and that encloses the projection rectangle, which is perpendicular to the cone axis. The angle between opposite sides of the viewing frustum is called the viewing angle. It is generally easier to let the viewing angle specify the angle for one dimension of the projection rectangle, and then to tailor the angle of the perpendicular angle of the viewing frustum to match the other dimension of the projection rectangle.

The greater the viewing angle, the greater the amount of perspective (wide-angle effect), and the lower the viewing angle, the lower the amount of perspective (telephoto effect). The viewing angle must reside in the range of 0 to pi, exclusive.

 The 3D Viewing Vector and Viewing Frustum

The angle from D to From to B is the horizontal viewing angle, and the angle from A to From to C is the vertical viewing angle.

To render a three-dimensional scene, we use these viewing parameters to project the scene to a two-dimensional rectangle, also known as the viewport. The viewport can be thought of as a window on the display screen between the eye (viewpoint) and the 3D scene. The scene is projected onto (or ``through'') this viewport, which then contains a two-dimensional projection of the three-dimensional scene.

Viewing in Four-Space

To construct a viewing model for four dimensions, the three-dimensional viewing model  is extended to four dimensions.

Three-dimensional viewing is the task of projecting the three-dimensional scene onto a two-dimensional rectangle. In the same manner, four-dimensional viewing is the process of projecting a 4D scene onto a 3D region, which can then be viewed with regular 3D rendering methods. The viewing parameters for the 4D to 3D projection are similar to those for 3D to 2D viewing.

     As in the 4D viewing model, we need to define the from-point. This is conceptually the same as the 3D from-point, except that the 4D from-point resides in four-space. Likewise, the to-point is a 4D point that specifies the point of interest in the 4D scene.

     The from-point and the to-point together define the line of sight for the 4D scene. The orientation of the image view is specified by the up-vector plus an additional vector called the over-vector. The over-vector accounts for the additional degree of freedom in four-space. Since the up-vector and over-vector specify the orientation of the viewer, the up-vector, over-vector and line of sight must all be linearly independent.

4D Viewing Vectors and Viewing Frustum

The viewing-angle is defined as for three-dimensional viewing, and is used to size one side of the projection-parallelepiped; the other two sides are sized to fit the dimensions of the projection-parallelepiped. For this work, all three dimensions of the projection parallelepiped are equal, so all three viewing angles are the same.

RAY TRACING ALGORITHM:

Raytracing solves several rendering problems in a straight-forward manner, including hidden surfaces, shadows, reflection, and refraction. In addition, raytracing is not restricted to rendering polygonal meshes; it can handle any object that can be interrogated to find the intersection point of a given ray with the surface of the object. This property is especially nice for rendering four-dimensional objects, since many N-dimensional objects can be easily described with implicit equations.

           A 2x2x2 4D Raytrace Grid

 

 Other benefits of raytracing extend quite easily to 4D. As in the 3D case, 4D raytracing handles simple shadows merely by checking to see which objects obscure each light source. Reflections and refractions are also easily generalized, particularly since the algorithms used to determine refracted and reflected rays use equivalent vector arithmetic.

VOLOCITY- A RENDERING SYSTEM

“New dimensions in High Performance Imaging

     Volocity is the realization of Improvision's objective to provide the scientist with an interactive volume visualization system that will run on a standard desktop computer. Volume interactivity is the key to providing the user with an enhanced perception of depth and realism. Interactivity also allows the scientist to rapidly explore and understand large volumes of data. The highly advanced technology developed exclusively by Improvision for rapid, interactive volume visualization of 3D and 4D volumes has received several awards for innovation and is the subject of worldwide patent applications.

     Volocity is the first true color 4D rendering system designed for biomedical imaging. It uses new highly advanced algorithims to provide high-speed, easy to use, interactive rendering of time resolved color 3D volumes. Volocity allows the user to visualize a 3D object and then observe and interact with it over time, for the first time providing scientists with a system to dynamically visualize both the structure and the purpose of biological structures.

Volocity Acquisition:

The Volocity Acquisition Module is designed for high performance acquisition of 3D sequences. It provides easy to use, high speed image capture capability and is compatible with a range of scientific grade cameras and microscopes.

Volocity Acquisition incorporates a unique parallel processing   and video streaming architecture. This technology is designed to acquire image sequences direct to hard disk at the maximum frame rate possible from each supported camera. The direct to disk streaming technology has the additional benefit of continuously saving acquired data.
Images are captured directly into an Image Sequence window and can be exported as QuickTime or AVI movies.
 Features

• Fast and highly interactive volume exploration in 3D and 4D.            

·         High speed parallel imaging and video streaming architecture.

• Fly-through rendering for visualizing events inside biological structures.  

·         Real time Auto Contrast for fast focusing and acquisition

• Intuitively easy to use for increased productivity.
• Object classification, measurement and tracking in all dimensions.
• High quality restoration of confocal and wide field microscope images.

Volocity Visualization

The Volocity Visualization Module provides an extensive range of visualization and publication features. The Volocity 3D view enables the user to interactively explore a 3D rendered object. Volocity Visualization also includes the Movie Sequencer, a unique authoring tool enabling the user to create and store a volume animation template. This template or Movie Sequence can be applied to any 3D Rendering View to create a series of pictures, for export as an AVI or QuickTime movie file. The Movie Sequencer is an easy to use feature for collection of visually unique animations, which clearly present the information of relevant scientific interest.

 Features:

• Interactive rendering of high resolution 3D and 4D volumes.
• Rapid publication of 3D and 4D movies as AVI, QT and QTVR files.
• A versatile range of 3D rendering modes for different sample types.
• Perspective rendering for enhanced realism.
• Fly-through rendering for visualizing events inside biological structures.
• Easy to use animation controls for playback of time resolved volumes.

Volocity Classification

     Volocity Classification is designed to identify, measure and track biological structures in 2D, 3D and 4D. This unique module incorporates innovative new classification technology for rapid identification and quantitation of populations of objects in 2D and 3D.The Classifier Module enables the user to 'train' Volocity to automatically identify specific biological structures. Complex classification protocols for detecting objects can be created and saved to a palette and then executed with a. Classifiers can be applied to a single 3D volume, to multi-channel 3D volumes and to time resolved volumes.

Features:

• Rapid classification and measurement of 2D, 3D and 4D images.
• Automatic tracking of objects in 2D and 3D.
• Overlay of measured objects on a 2D image or 3D volume.
• Comprehensive range of intensity, morphological and volume measurements.
• Measurements from multiple channels for channel comparison and colocalization.
• Data export to industry standard spreadsheets.

Volocity Restoration

     Volocity Restoration includes restoration algorithms and measured or calculated PSF generation options for confocal and wide field microscopes. The Volocity restoration algorithms are designed for rapid, high quality restoration of 4D and 3D volumes and for accurate comparison and quantitation of time resolved changes.

The Iterative Restoration algorithm is an award winning restoration algorithm developed by Improvision from published Maximum Entropy techniques. The Iterative Restoration feature is an exceptional technique for eliminating both noise and blur to produce superior restoration results and significant improvement in resolution in XY and Z.

The Fast Restoration algorithm is an ultra-fast routine developed by Improvision. This algorithm uniquely uses every voxel in the volume in a single pass process to improve both the visual quality and the precision of the result. This feature is extremely fast to compute and produces superior results when viewed in XY.

 Features

• Iterative Restoration for improvement of XY and Z resolution.
• Fast Restoration for improvement of XY resolution.
• Confocal and Wide Field PSF Generator.
• Tools for illumination correction.
• Measured PSF Generator for confocal and wide field images.
• Batch processing of 3D sequences.

4D VISUALIZATION IN LIVING CELLS:

     Due to recent developments in genetic engineering, cell biologists are now able to obtain cell-lines expressing fusion proteins constituted with a protein of interest and with an auto fluorescent protein: GFP. By observing such living cells with a confocal microscope, it is possible to study their space-time behavior.

     Because they correspond to the dimensionality of what the physical reality is in contrast with 2D or/and 3D images which are 'only' reductions by projection or time fixation, 4D images provide an integral approach to the quantitative analysis of motion

and deformation in a real tri-dimensional world over the time.

OBJECTIVES:

o   To easily visualize the evolution of the structures in 3D during all the steps of the experiment.

o   To be able to select one given structure in order to study its spatial behavior relatively to others (fusion, separation.)

o   To measure parameters such as: • shape, volume, number, relative spatial

                        localization for each time point, • trajectories and speed of displacement   

                        for each structure either alone or relatively to the others                     

o   To cumulate the data of several structures within many cells and in different experimental conditions in order to model their behavior.

Volumes Slicing and Projection: Reduction from 4D to 3D:

     The slicing operation consists in choosing one of the parameters corresponding to the dimensions {x, y, z, t}, and giving it a constant value. The most usual choice is to consider the data for a fixed value of the time t. In the projection operation, for each {x, y, z} volume at a given time, a single 2D image is obtained by integrating the data along

an axis, the z-axis for example. This can be done using a Maximum Intensity Projection algorithm. Then, all these projection images are used to build a new volume (x, y, t), as

illustrated in Figure. It has the advantage under certain conditions

 to correctly illustrate topological changes over time.

     As after these operations we obtain “classical” 3D volumes, we can apply 3D visualization methods, like iso surfacing, direct volume rendering and others to compare their effectiveness to correctly analyze and interpret all the information contained in the

series.

Space-Time Extraction

Here the idea is to model the objects of interest and their evolution, with a 4D tool that is also suited for visualization and quantification.

Multi-Level Analysis:

The reconstruction of one object in the 4D data can be applied for very kind of objects, even if one is included inside another one.

Figure shows two levels corresponding to the nucleolus and UBF-GFP spots. For example, one can compute the evolution of both the nucleoli and the UBF-GFP proteins. If the nucleoli are moving fast, the movements of the UPF-GFP proteins must significantly be corrected, to take it into account.

IMAGE PROCESSING TECHNIQUE:

Parameter Gluing:

     A main scientific approach is to reduce the complexity of a phenomenon by choosing a parameter of the system, setting it to a defined value (gluing it!), and analyze how the other parameters evolve.

The trajectories can be represented in three different modes, according to the specifications of the user. Their normal representation is via a set of cylinders showing the evolution of the objects, with small sphere for the topological breaks. This representation is enhanced by modifying the radius of cylinders according to the volume of the objects (see Figure ).

      Two proteins merged into one

Integration of time to a spatial dimension:

In the last visualization mode, time is integrated to a spatial dimension, leading for example to a {x, y, z+t} 3D representation (see Figure). This approach has the advantage of better presenting the data variations according to time. It has the advantage not to alter the data during the processing. It produces a representation of the evolution of the center of mass of the object.

                                                                                                                 

Space-Time deformable model:

It enables the representation of the evolution of the objects, and different visualization modes.

Figure shows that the evolution of the spots of the upper part of the image is not easy to qualify. Sometimes false merges occur: when two spots are becoming closer, they may be merged by the deformable model. The resolution of the model is thus important for the reconstruction.

WORKSTATION FOR ACQUISITION, RECONSTRUCTION AND VISUALIZATION OF 4D IMAGES OF HEART.

Objectives
A Workstation is developed for the acquisition, reconstruction, processing and visualization of 4D images of the heart. These images are obtained from two-dimensional echocardiography equipments using the method of transtoracic rotational sweep.

Methodology
One important step to the reconstruction of 4D images of the heart, is to build an echocardiography database. This is generally obtained by scanning the heart with the ultrasound sheaf; In this work, the method of Transtorasic Rotational Fan is used. In this method, the transducer is rotated on its longitudinal axis

 Transtoracic rotational sampling, apical see. a) Position and movement of the transducer. b) Two-dimensional Echo obtained on the slice plane 1. c) Volume swept by the ultrasonic sheaf.

     To obtain a sequence of 4D-data of the heart, a commercial two-dimensional echocardiography equipment is used, but the processes involved are complicated. This is because the volumetric images should be generated during the heart cycle, they should be obtained starting from 2D echoes captured in different times and in different planes of acquisition. Additionally, there are diverse anatomical events that generate spatial noise and the elimination of small anatomical detail during the 4D-data acquisition can also produce errors. These inaccuracies can degenerate the reconstructed image. The acquisition of data is synchronized with the breathing rhythm and the heart rhythm; the manipulation of the ultrasonic transducer is realized by a motorized servo-mechanism controlled by computer. A graphic processing system  is also used to allow the control of the whole station, the processing and visualization of the 4D-database. The figure 2 shows a scheme of the designed Workstation.

 A scheme of the designed Workstation. The ultrasonic test is realized in real time using a two-dimensional echo graphic ESOATE PARTNER model AU-3 with a sectoral multi-frequency transducer of 3.5/5.0 MHz.

RESULT

 Three-dimensional Image of an aortic bi-valve. To the left in dyastole and to the right in systole. It is appreciated an augment of the border of the valves and the complete opening of the same ones.

     The obtained results are highly satisfactory. The 4D images of the heart obtained using the Workstation are of high quality; it shows the details of the heart cavities and the valvular structure. From the clinical point of view this acquires great importance since the medical exam is simple and non-invasive. Moreover, it doesnít need sophisticated equipments, and it can be installed in a consultory. Additionally, objective data from the heart anatomy is obtained. These images can be a useful guide for the cardiovascular surgeon.

4D IMAGE WARPING FOR MEASUREMENT OF LONGITUDINAL BRAIN CHANGES:

     For robustly measuring temporal morphological brain changes, a 4D image warping mechanism can be used. Longitudinal stability is achieved by considering all temporal MR images of an individual simultaneously in image warping, rather than by individually warping a 3D template to an individual, or by warping the images of one time-point to those of another time-point. Moreover, image features that are consistently recognized in all time-points guide the warping procedure, whereas spurious features that appear inconsistently at different time-points are eliminated. This deformation strategy significantly improves robustness in detecting anatomical correspondences, thereby producing smooth and accurate estimations of longitudinal changes. The experimental results show the significant improvement of 4D warping method over previous 3D warping method in measuring subtle longitudinal changes of brain structures.

METHOD:

4D-HAMMER, involves the following two steps:

(1) Rigid alignment of 3D images of a given subject acquired at different time points, in order to produce a 4D image. 3D-HAMMER is employed to establish the correspondences between neighboring 3D images, and then align one image (time t) to its previous-time image (t-1) by a rigid transformation calculated from the established

correspondences.

(2) Hierarchical deformation of the 4D atlas to the 4D subject images, via a hierarchical attribute-based matching method. Initially, the deformation of the atlas is influenced primarily by voxels with distinctive attribute vectors, thereby minimizing the chances of poor matches and also reducing computational burden. As the deformation proceeds, voxels with less distinctive attribute vectors gradually gain influence over the deformation.

 

Comparing the performances of 4D- and 3D- HAMMER in estimating the longitudinal changes of from a subject.

MED-SANARE

Medical Diagnosis Support System within a Dynamic
Augmented Reality Environment

     Advanced medical imaging technology allows the acquisition of high resolved 3D images over time i.e.4D images of the beating heart. 4D visualization and computer supported precise measurement of medical indicators (ventricle volume, ejection fraction, wall motion etc.) have the high potential to greatly simplify understanding of the morphology and dynamics of heart cavities, simultaneously reduce the possibility of a false diagnosis. 4D visualization aims at providing all information conveniently in single, stereo, or interactively rotating animated views.

     The goal of the 2nd year of the Med-SANARE project is twofold. On one hand a virtual table metaphor will be utilized to set up a visionary high-end cardiac diagnosis demonstrator for educational purpose that makes use of augmented reality (AR) techniques. On the other hand a Cardiac Station will be implemented as functional reduced solution that supports image evaluation making use of standard PC-based technology. The functionality offered will be sufficient to successfully perform the tasks required by the diagnostic procedure.

     For both systems realistic and detailed modeling and visualization plays a crucial role.

 Modeling/Visualization Pipeline:

  

      The data is either visualized without any preprocessing applying direct volume rendering, or in the first step segmented by application of semi-automatic 2D/3D segmentation methods. A subsequent triangulation process transforms the result into hardware renderable polygonal surfaces that can also be tracked over the temporal sequence. Finally the time-variant model is visualized by application of advanced 5D visualization methods.

BIBLIOGRAPHY

W. de Leeuw and R. van Liere. Case Study: Comparing Two Methods for Filtering External Motion in 4D Confocal Microscopy Data. Joint Eurographics .

W.E. Lorensen and H.E. Cline. Marching cubes: a high resolution 3D surface construction algorithm. Computer Graphics (Siggraph’87 Proceedings), 21(4), pp163-169.

M. Levoy. Display of surfaces from volume data. Computer Graphics & Application.

P. Lacroute and M. Levoy. Fast volume rendering using shear-warp factorization of the viewing transform. Computer Graphics (Siggraph’94 Proceedings).

B. Cabral, N. Cam and J. Foran. Accelerated volume rendering and tomographic reconstruction using texture mapping hardware.

O. Wilson, A.V. Gelder and J. Wilhems. Direct volume rendering via 3D textures. Tech. Rep. UCSC-CRL-94-19, University of California at Santa Cruz.

C. Rezk-Salama, K. Engel, M. Bauer, G. Greiner and T. Ertl. Interactive volume rendering on standard PC graphics hardware using multi-textures and multi-stage rasterization. Siggraph & Eurographics Workshop on Graphics Hardware 2000, 2000.

J-O. Lachaud and A. Montanvert. Deformable Meshes with Automated Topology Changes for Coarse-to-fine 3D SurfaceExtraction. Medical Image Analysis.