Overview

This appendix briefly reviews developments in the medical applications of virtual environments technology. While it puts the research described in this thesis in context, this appendix is independent of the rest of the text.

There is considerable disagreement over what should and should not be covered by the term ``virtual environments.'' A definition in terms of interface considerations might be based on interactive three-dimensional displays, whether visual, aural, tactile, force-feedback, or some combination of the above.

Another definition, coming from a psychological viewpoint, centers around the idea of ``presence'': a virtual environment is one which substantially convinces the participant that he/she is ``in'' the environment being presented.

More difficult than defining what virtual environments are is defining what they are not. The term usually excludes interactive, screen-based computer graphics with a 2-D mouse. But what about the above with a 3-D mouse? Or a workstation which provides a 3-D display using (for instance) shutter-glasses, but which provides only a limited range of movement? What about virtual objects overlaid on a natural scene?

As with any definition, the goal should be to define the term in such a way as to include a significant self-related body of thought and work, while excluding unrelated or tangentially related material. For the purposes of this appendix, it seems best to use a loose definition in terms of intent: ``virtual environments'' are displays which attempt to present information in ways which are natural or intuitive to human users, and which make use of natural human sensory and interaction modalities.

Thus, we do not arbitrarily include or exclude traditional interactive computer graphics from this definition; but graphical displays which are truly 3-D and which have 3-D interaction devices are in terms of the above definition ``better'' virtual environments, in that they come closer to natural human sensory and interaction modalities.

As a practical matter, ordinary computer graphics as applied to medicine will not be discussed in this appendix, simply because the topic is too huge to be treated well in a limited amount of space. Instead, the focus is on ``true'' 3-D displays and interactions.

While a great deal of work has been done on the application of virtual environments to medical problems [61,50,59], most is of a preliminary nature.

One possible breakdown of medical applications of virtual environments is into visualization (or, more generally, sensing) tools, simulation tools (either for training or pre-operative planning), and assistive technology tools. Visualization tools are systems which attempt to provide medical data in a readily understandable way. Simulation tools are aimed at teaching procedures, most commonly (at present) surgical procedures. Visualization and simulation tools are intended primarily for the care-giver; by contrast, assistive technology tools attempt to provide new abilities to clinical patients.

Current computing limitations force a trade-off between level of detail and level of interaction. Visualization tools tend towards a very high level of detail but not necessarily much interaction, whereas simulators tend to emphasize interactivity at the expense of detail.

We will consider each of the three areas below. In terms of the above division, the work described in this thesis falls into ``assistive technologies''.