Body Mapping Tools
[NB: This demo will not run under all security settings, browser versions, or operating systems. A recent Windows security patch has decided that this little interface is, indeed, an enemy of the state. Therefore, I have included some still images. They don’t provide a feeling of the pan and zoom workflow but you do get a feeling of the appreciable scale and functionality available from a hierarchy that has just a few levels.]
One difficulty associated with tracking lesions on the skin over time is the ability to repeatedly match related diagnostic information with regions of interest anywhere on the 3D body surface. The difficulty of this task is compounded by the need to unambiguously determine the spatial position of any of several lesions that often may be in close proximity so that all associated diagnostic quality images, histology, or other lesion-specific data required to make clinical decisions can be appropriately retrieved. To explore the ergonomic trades in such a task, a version of MGI’s Zoom interface was adapted to incorporate the front and back views of a mannequin. The Zoom interface allows images of different scales to be linked together and explored. For applications that require a relatively dense mosaic of regions, it is possible to employ tool tips that can be used to identify each region upon a mouse over. This configuration doesn’t solve all of the problems of body mapping; however, it does provide a platform to explore some of the new approaches to accessibility and workflow issues.
The combination of a slightly rough surface covered by a realistically colored, matte skin tone and realistic scalp hair made Paolo2 a good imaging testbed. Unfortunately, he is apparently no longer available at Mannequinland and since prior history has shown that the half-life of intact mannequins in university laboratories is quite short, this is a concern. Issues of ephemerality notwithstanding, a hierarchy of four image scales was sufficient to span the range required to capture the 1.8 meter mannequin down to a 1 centimeter macro image, a 180:1 magnification ratio. The 4 micrometer pixel sampling afforded by the macro image results in a linear partitioning of the body’s height to approximately 1 part in 450,000, measured by projecting pixels onto the skin surface.
Based on this pilot data, a second generation prototype is being developed. This new interface addresses the difficulty with tracking multiple lesions. Version 2 will provide a method of capturing and organizing the required information in a manner that is logical, accessible, and rapid. It is expected that the new procedure will acquire body surface images with a procedure that will require only moderate training and that should fit well with existing staff capabilities and clinic workflow. The interface addresses another body mapping problem by providing an improved method of cataloging the images for future reference and for comparison across mappings.