Work at DIGITAL's (DEC) Cambridge Research Lab (CRL)

...under construction...

VDI: a Visual Debugging Interface

Higher-level image interpretatin systems typically consist of many tightly-coupled modules, each related to a particular aspect of the interpretation task. When such a system produces wrong results, programmers generally spend a lot of time writing special-purpose display routines to facilitate investigating the problem. This paper presents an attempt at unifying such efforts by merging graphics programming with debugging technology. Several generalizations are necessary to provide useful general-purpos display mechanisms: 1) The mechanisms need to be highly interactive such taht programmers can select and change display preferences at debugging time. 2) The display routines should not be restricted to traditional, two-dimensional images. 3) The programmer must be able to visualize the data-dependent relationships between modules. VDI is a general-purpose Visual Debugging Interface designed to provide these generalizations. It is highly interactive, operates on arrays of arbitrary dimensionality, and provides mechanisms to describe dependencies between the results generated by various interdependent modules.

Videos

• Basic functionality (1 MB) (22 MB)
• Link between a color image and a color histogram (2.5 MB) (20 MB)
• Use of VDI to analyze and explore the use of various heuristics and algorithms to separate color images into diffuse and specular relflection components (13.5 MB) (134 MB) (see also Ph.D. thesis at CMU)

TDE: a workbench for Tele-collaborative Data Exploration

We have developed an environment for telecollaborative data exploration. It provides the following capabilities essential to data exploration: (1) Users can probe the data, defining regions of interest with arbitrary shapes. (2) The selected data can be transformed and displayed in many different ways. (3) Linked cursors can be established between several windows showing data sets with arbitrary relationships. (4) Data can be displayed on any screen across a computer network, allowing for telecollaboration arrangements with linked cursors around the world. (5) Our system is user-extensible, allowing programmers to change any component of it while keeping the remaining functionality. We demonstrate how the system can be used in several applications, such as biomedical imaging, robotics, and wood classification.

Videos

• Basic functionality (5 MB)
• Data probing (3 MB)
• Cursor linking (2 MB)
• Cropping (5.5 MB)

Applications (TDE)

Statistical analysis tools to quantify tissue vascularity in images from the thyroid of a rat

We have developed a data exploration system which provides tools which establish linked cursors between several windows. The system extends AVS, a commercial visualization system, to provide data probing and cursor linking facilities which have proven invaluable for tele-medicine and for statistical data analysis involving image data and histograms.

Shown is a serial section from the thyroid of the rat. The tissue has been stained so that blood capillaries appear blueish.

In the movie, the data is shown in two windows that are used for tele-collaboration. The cursors of the two windows are cross-linked such that we can independently annotate the data in each window using different colors. The annotations are shown in both windows. Using the built-in networking facilities of X-Windows, we send one of the windows to a communication partner across our computer network. Users can customize the individual views. For example, they can resize the window or recolor it. This technology can be used to share data between any X-based displays around the world. We have established such connections across the United States as well as between the US and Sweden.

Our system combines tele-collaboration with extensive data exploration tools. Users can interactively establish linked cursors between many different data sets. For example, we can link images with histograms or scatter plots.

For the thyroid image, researchers are interested in developing methods to automatically quantify tissue vascularity by detecting the blue pixels in the image. When the user outlines an area of the histogram, our system highlights all pixels with such colors. Such cross-linking between the image and the histogram is essential to the statistical analysis of image data.

Our system uses a visual programming interface with which users interactively set up and modify their data exploration environment. Because of this interactive flexibility, the system is amenable to many forms of data exploration in a tele-collaborative environment.

Videos

• Demo (18 MB)

Tools to register and segment soft tissue from medical images and visualize it three-dimensionally

Numerous techniques have been developed for biomedical image segmentation, but most methods do not perform well on a broad variety of data. We present an interactive approach to the segmentation of medical images. This approach is based on interactive deformable contours and has broad applicability to soft tissue segmentation. We describe a system, implemented in AVS, that we have applied to the segmentation of nerve cells and a human embryo heart from serial tissue sections, and the segmentation of lymph nodes from CT data.

Videos

• Demo (30 MB)

Tools to explore the boundary of a lymph node

Average intensities are one criterion for distinguishing stained cancerous lymph nodes from healthy ones. We are studying the use of a physically based segmentation algorithm called SNAKES to outline lymph nodes in CT-scan slices and to determine their average intensity.

With the SNAKES algorithm we can interactively sketch an outline. The algorithm then smoothes the contour and adjusts its contour to coincide with places of maximal response from an edge detector. Users can adjust the outline interactively as necessary.

We use a suite of interactive data exploration tools to study how the generated outline depends on the edge detector.

Alltogether, the visualizations illustrate that the intended computation of the average intensity value of the lymph data is critically dependent on the boundary detection heuristics. Our visualization tools can be used to explore a suite of different edge operators approximating the first or second derivative of the image data to determine which arrangement suits the lymph data best and why.

Videos

• Demo (16 MB)

PerfVis: A Performance Visualizer for High Performance Fortran Programs on Workstation Farms

Writing efficient code for parallel processors is still a rather complex and little understood task. Tools to gather and analyze performance information during program execution are essential for understanding the reasons for inefficient executions. Current performance visualization systems provide only a limited set of capabilities, hardwired into huge, monolithic programs. Programmers need a very flexible environment in which they can mix and match different performance visualization tools. Research in scientific visualization has developed several environments to visualize, explore and analyze large quantities of data. Reusing these capabilities to visualize MIMD performance data significantly helps in the development of a performance profiler. However, scientific visualization methods have to be altered appropriately to address the non-geometric nature of performance data. This paper presents a profiler, PerfVisS, which builds upon existing Telecollaborative Data Exploration (TDE) technology. PerfVisS provides several linked views of the performance data, both for gaining a general overview of the program performance and for detailed data and code inspection. It also allows users to adapt the system to their own viewing preferences, using the AVS visual programming interface. Furthermore, it provides tools for telecollaborative performance tuning, allowing programmers to share their views with colleagues at remote sites. PerfVisS runs on Alpha AXP(TM) workstation farms but could be used as well on other similar MIMD architectures such as symmetric multiprocessors (SMPs).

Publications

1996
	G. Klinker Interactive Visualization Environment with Improved Visual Programming Interface US Patent 5,590,271. (bib)
1995
	I. Carlbom, T. Kapur, G. Klinker, D. Terzopoulos, L. Thurfjell General-Purpose Soft Tissue Segmentation from Medical Images Theory and Applications of Image Processing II - Selected contributions of the 9. Scandinavian Conference on Image Analysis (SCIA), Swedish Society for Automated Image Analysis, ed. G. Borgefors, World Scientific, July 1995. (Also in the Proc. of SCIA, Uppsala, Sweden, June 1995, S. 905-912). (bib)
	G. Klinker, I. Carlbom, W. Hsu, D. Terzopoulos Biomedical Data Exploration Meets Telecollaboration Proc. 1rst International Conference on Computer Vision, Virtual Reality and Robotics in Medicine (CVRMed'95), Nice, France, Apr. 1995, Springer Verlag Lecture Notes in Computer Science, Vol. 905 (ed. N. Ayache), pp. 84-91. (Extended version with color figures can be found in CRL TR 94/6). (bib)
1994
	G. Klinker, I. Carlbom, W. Hsu, D. Terzopoulos Scientific Data Exploration Meets Telecollaboration Technical report 94/6, Digital Equipment Corporation, Cambridge Research Lab (DEC-CRL), One Kendall Square, Cambridge, MA, June 1994. (Extended version of paper at CVRMed 1995). (bib)
	G. Klinker, I-Yu Chen PerfVisS: A Performance Visualizer for High Performance Fortran Programs on Workstation Farms Technical report 94/4, Digital Equipment Corporation, Cambridge Research Lab (DEC-CRL), One Kendall Square, Cambridge, MA, June 1994. (bib)
	G. Klinker Coroutine Synchronization in AVS 3rd International AVS User Group Conference (AVS'94), Boston, MA, May 2-4, 1994. (bib)
1993
	G. Klinker An Environment for Telecollaborative Data Exploration IEEE Visualization'93, San Jose, CA, Okt. 1993, pp. 110-118. (bib)
	G. Klinker Interactive Data Exploration and Telecollaboration in Biomedicine Using AVS 2nd International AVS User Group Conference (AVS'93), Walt Disney Dolphin, FL, May 24-26, 1993. (bib)
	G. Klinker An Environment for Empirical Data Interpretation White paper, Digital Equipment Corporation, Cambridge Research Lab (DEC-CRL), One Kendall Square, Cambridge, MA, January 1993. (Detailed description of the fundamental concepts presented in later publications). (bib)
1992
	I. Carlbom, W. Hsu, G. Klinker, R. Szeliski, K. Waters, M. Doyle, J. Gettys, K.M. Harris, T.M. Levergood, R. Palmer, L. Palmer, M. Picart, D. Terzopoulos, D. Tonnesen, M. Vannier, G. Wallace Modeling and Analysis of Empirical Data in Collaborative Environments Communications of the ACM (CACM), Special Issue on the Siggraph'92 Showcase, Vol. 35, Number 6, June, 1992. (bib)
	G. Klinker VDI - A Visual Debugging Interface for Image Interpretation and Other Applications in Advances in Scientific Visualization, F.H. Post and A.J.S. Hin (eds.), Springer Verlag, 1992. (Selected paper from the Eurographics Workshop on Visualization in Scientific Computing, Delft, Netherlands, April 1991.) (bib)
1991
	G. Klinker We Need Interactive Data Interpretation Rather Than interactive Data Visualization Position Paper for the Workshop on Scientific Visualization Environments, IEEE Visualization'91, Okt. 1991. (bib)
	G. Klinker VDI - A Visual Debugging Interface for Image Interpretation and Other Applications Technical report 91/2, Digital Equipment Corporation, Cambridge Research Lab (DEC-CRL), One Kendall Square, Cambridge, MA, March 1991. (bib)