Mike Bailey

Rapid prototyping to create vascular replicas from CT scan data: making tools to teach, rehearse, and choose treatment strategies.

Our goal was to develop and prove the accuracy of a system that would allow us to re‐create live patient arterial pathology. Anatomically accurate replicas of blood vessels could allow physicians to teach and practice dangerous interventional techniques and might also be used to gather basic physiologic information. The preparation of replicas has, until now, depended on acquisition of fresh cadaver material. Using rapid prototyping, it should be able to replicate vascular pathology in a live patient. We obtained CT angiographic scan data from two patients with known arterial abnormalities. We took such data and, using proprietary software, created a 3D replica using a commercially available rapid prototyping machine. From the prototypes, using a lost wax technique, we created vessel replicas, placed those replicas in the CT scanner, then compared those images with the original scans. Comparison of the images made directly from the patient and from the replica showed that with each step, the relationships were maintained, remaining within 3% of the original, but some smoothing occurred in the final computer manipulation. From routinely obtainable CT angiographic data, it is possible to create accurate replicas of human vascular pathology with the aid of commercially available stereolithography equipment. Visual analysis of the images appeared to be as important as the measurements. With 64 and 128 slice detector scanners becoming available, acquisition times fall enough that we should be able to model rapidly moving structures such as the aortic root.

Tele-Manufacturing: rapid prototyping on the Internet

Discusses how a research and development project called the Tele-Manufacturing Facility is creating an automated rapid prototyping capability on the Internet. We are doing the necessary research and development to make the capability viable for engineers and scientists to use over long distances. Our view of the engineering concept design-production process is given. This model clearly shows the iterative nature of the design process. >

VizWiz: a Java applet for interactive 3D scientific visualization on the web

VizWiz is a Java applet that provides basic interactive scientific visualization functionality, such as isosurfaces, cutting planes, and elevation plots, for 2D and 3D datasets that can be loaded into the applet by the user via, the applets Web server. VizWiz is unique in that it is a completely platform independent scientific visualization tool, and is usable over the Web, without being manually downloaded or installed. Its 3D graphics are implemented using only the Java AWT API, making them portable across all Java supporting platforms. The paper describes the implementation of VizWiz, including design tradeoffs. Graphics performance figures are provided for a number of different platforms. A solution to the problem of uploading user data files into a Java applet, working around security limitations, is demonstrated. The lessons learned from this project are discussed.

The use of solid physical models for the study of macromolecular assembly

The use of modern technology in the construction of accurate solid macromolecular models based on atomic coordinates and electron density functions has led us to re-examine the usefulness of physical models as tools for understanding molecular assembly and for designing detailed experimental and computational studies of the assembly process. Recent developments include the construction of new models, which have provided insights into the assembly of viruses and light harvesting complexes.

Interacting with direct volume rendering

Visualization researchers at the San Diego Supercomputer Center have written a program called Volume Explorer, or vx, which performs interactive direct volume rendering on the Volume Pro board. Using OpenGL, the GL Utility Toolkit and the Micro User Interface, vx generates the graphics and facilitates user interaction around the Volume Pro card. Regardless of the method, my colleagues and I have become very big fans of interactive direct volume rendering. This article shows some of our preliminary work with vx. On the screen, you can change and manipulate images at update rates of more than 30 frames per second.

Stereoscopic Evaluation of Fetal Bony Structures

The purpose of this study was to assess the performance of stereoscopic compared with conventional viewing of 3‐dimensional ultrasound (3DUS) data for evaluation of fetal bony structures.

Lessons from scene graphs: using scene graphs to teach hierarchical modeling

Abstract The scene graph, as defined in VRML and Java3D, is a powerful tool for modeling a scene. The ideas contained in the scene graph are fundamental principles in modeling. They give beginning computer graphics students the tools to understand and apply the techniques of hierarchical modeling in scene design and can be directly applied to graphics programming in several graphics APIs, including OpenGL. This note outlines the approach to modeling with scene graphs and describes how students in a first computer graphics course can build their modeling designs with this approach.

Augment Your Reality

This article features some of the latest advances and applications in computer graphics technology.

Using GPU Shaders for Visualization

GPU shaders seem used mostly for gaming and other forms of entertainment and simulation. But they have less-obvious visualization uses, for the same reasons that interest the gaming community: improved appearance and performance. This column looks at the use of shaders and the OpenGL shading language (GLSL) in two common visualization applications: point clouds and contour cutting planes.

A hands-on environment for teaching GPU programming

GPU programming is fast becoming an essential skill for computer graphics students. It has immediate application in all areas of graphics including science, engineering, art, animation, and gaming. Because it is new, experience with teaching GPU programming is scarce. This paper describes the teaching of a GPU programming course with a hands-on program called glman. glman allows students to create a shader scene description file which not only creates the 3D scene, but creates an interactive user interface to adjust shader parameters. Our experience in an experimental class taught in Spring 2006 is that glman is flexible enough to donstrate and experiment with many shader concepts, and creates a fast and fun learning curve for the students.