Demonstration
Can a Virtual Cadaver Be Developed From the Visible Human Male?
Don R. Hilbelink, Ph.D. University of South Florida
Abstract:
Introduction
Although a computer based virtual cadaver can never fully replace the experience of doing an actual cadaver dissection there are a variety of roles that such a program could play in medical education. Many anatomists believe that for a virtual cadaver to be accepted for incorporation into a medical gross anatomy course it optimally should contain a level of detail comparable to a classic cadaver dissection. The Visible Human Project of the National Library of Medicine (NLM) has provided a resource from which a virtual cadaver could possibly be developed. This study was conducted to determine the feasibility of success of such a project and to develop techniques for its implementation.
Methods
Anatomical structure lists were established for the neck region of the body representative of two levels of detail. The first level approximated that of a typical medical gross anatomy dissection course whereas the second contained all of the neck anatomical structures listed in the Terminologia Anatomica. A full resolution (pixels: 0.33 mm) set of images of the VHM was obtained from Gold Standard Multimedia. Images inclusive of the neck region of the VHM were systematically examined to identify each of the structures contained on our lists of anatomical structures. An extensive number of neck structures could not be resolved at the 0.33 mm pixel image resolution. A set of higher resolution VHM images (pixels = 0.15 mm), produced by scanning of the original 70 mm color negatives of the VHM sections at a 6,000 by 4,000 pixel resolution, was obtained from the NLM. These images were examined in a fashion identical to our previous approach to determine if additional structures could be resolved and segment
Results
A list of anatomical structures of the human neck representative of two distinct levels of detail has been produced. Techniques have been developed to permit registration and segmentation of high-resolution image sets of the VHM produced by scanning of he original 70-mm color negatives at the NLM. The initial observation, based on analysis of the neck region of the VHM image data set, is that it will be possible to produce a virtual cadaver program with an anatomical detail level, equivalent to or exceeding that, of the typical medical gross anatomy course taught in the USA. If any limitation to success of such a project currently exists, it is the fixed slice thickness (1-mm) of the VHM images, that is not altered with increased image resolution. This problem appears manageable if one is permitted a limited degree of "creative anatomical license" with development of the segmentation set.
Benefit in Attending Session:
The Visible Human Project has been the center of much attention. The actual breath of the data set has not yet been explored. This presentation will help put into perspective the possibility for high-end medical education use for this data set.
Don R. Hilbelink, Ph.D. MDC Box 6 12901 Bruce B. Downs Blvd. Tampa, Florida 33612 813-974-9483 Fax Number: 813-974-2958 dhilbeli@com1.med.usf.edu