Professor KW Brodlie

BSc (Edinburgh); MSc (Dundee); PhD (Dundee); AFIMA; FBCS; CEng; CMath; CITP

e-mail: k.w.brodlie@leeds.ac.uk tel: +44 (0)113 3435484 fax: +44 (0)113 3435468 room: E C Stoner Building 7.24 personal web address: http://www.comp.leeds.ac.uk/kwb/

Position

Professor of Visualization

Roles

Director of lab: Visualisation 6.01

Multidisciplinary Informatics Research Stream Lead

VVR Research Group Leader

Research activities

Ken's main research interest is in scientific visualization, an activity within the Scientific Computing and Visualization research group. This interest began in the 1970s with an involvement in the NAG Graphics Library, and he now leads a research group which researches in three main areas: visualization systems, visualization algorithms and more recently virtual environments.  The group includes Dr Jason Wood and a large number of research students, past and present; and owes much to collaboration with Dr Helen Wright, now at University of Hull.

Visualization systems: Our work aims to extend existing visualization systems in three directions

Problem-solving environments: This couples computation and visualization in a larger working environment - this work began with the GRASPARC project where a 'history tree' was used as a metaphor for the exploratory process of scientific investigation; and has continued more recently with the DIVA project where the exploration of high-dimensional parameter spaces was studied.

Collaborative visualization: This extends visualization systems from single-user to multi-user mode, allowing real-time synchronous collaboration.  We have developed the COVISA extension of IRIS Explorer, the world's first commercially available collaborative visualization system.

Web-based visualization: This investigates the different ways of providing visualization services over the Web. We are studying both server-based and client-based architectures.

Visualization algorithms: Algorithms essentially transform discrete numerical data into geometric models. Our work aims to improve the robustness of these algorithms in a number of ways:

Incorporating constraints: This allows the addition of constraint information (such as the model must be positive everywhere) into interpolation algorithms.

Ensuring visual continuity: This studies the effect of changes of data on the resulting geometric model, and has led to the development of a new, more accurate version of the Marching Cubes isosurfacing algorithm.

Understanding errors: This studies the possible errors created when a visualization algorithm is executed - for example, in generating a particle trace in flow visualization.

Virtual environments: Our work has focussed on virtual surgery and has two themes:

Web-based Virtual Surgery: This studies the architectures and algorithms suitable for providing surgical training via the Web. This is in collaboration with surgeons and radiologists at the Leeds Teaching Hospitals.

Reconstruction of Circle of Willis: This is looking at the reconstruction from MRA data of the artery structure in the Circle of Willis, in such a way that the model can be used in CFD analysis of the blood flow. This is in collaboration with the School of Mechanical Engineering.

Research students

Professor Brodlie is currently supervising the following research student(s):

Selected publications

or you can display a fuller, categorised list

Staff list