The Vesalius project not only relies on the Visible Human Data material from the U.S. National Library of Medicine (NLM) for the 3D image gallery but also builds on NLM's Unified Medical Language System (UMLS) standard vocabulary tool to design the knowledge base. The team developed its own 3D visualisation packages to process the Visible Human Data and to acquire a 3D map of the human body, in which each visible structure could be surface-based and textured with the true colour of its tissue. In addition, a query language was generated linking to the underlying ontology to browse the 3D map of the human anatomy and produce meaningful views for given applications.
Over the past few years, medical students at Columbia University were given a chance to try out the assets of the Vesalius 3D gallery and knowledge base in a variety of anatomy course applications. The neuro-anatomy curriculum for instance, provides the student with rapid access to static 2D images and 3D interactive graphics and animations. Also, clear learning goals have been defined to stimulate the integration of diverse sources of data rather than the accumulation of isolated facts. The course material is available on a central server but is being delivered over the Internet to enable a distributed multi-user access.
A second programme is dedicated to the human skull, featuring each cranial bone and showing how the different bones are integrated to form the skull. This application is highly interactive, using an "exploded" skull built from the actual photographs of the whole skull and its subparts. In a very clever way, the student can play with the parts, removing and putting them together again, like a puzzle. The curriculum equally incorporates the nervous system and demonstrates the complete pathways for all of the components of the cranial nerves. In this application as well as in others, the Versalius database is connected via interactive menus to information on physiology, neuro-anatomy, microscopic anatomy, and embryology, as well as to clinical correlations, pathology, and physical diagnosis.
A third example is the design of a three-dimensional interactive visualisation of the male pelvic anatomy for application as a tool to outline the uro-genital structures without the artifact of dissection. Three-dimensional voxel-based models are extracted from 2D data to build 3D surface-defined wrappings for spatial illustration of structures involved in prostate-related pathologies. The programme provides tools for the removal and replacement of various pelvic structures, leading to a greater understanding of relationships in this region. At this moment, the Vesalius team still continues to focus on improving 3D visualisation tools to help create anatomical models that can serve multiple purposes in teaching, research, and the commercial world. More details are available at the Web site of the Vesalius project.