Artifacts and illusions in surface and volume rendering

Surface and volume rendering from stacks of image slice data are used widely in both research and clinical environments. Well known rendering techniques implemented on high performance workstations present visualizations of multiple structures for diagnostic or scientific purposes. Although interpretation of these renderings can be more intuitive than sequential slice interpretation, both surface rendering and volume rendering make modeling assumptions, at least implicitly, which allow some parts of the image volume to be ignored or demphasized while other parts are highlighted. Three dimensional visualization methods have greatly increased the options for presenting the relevant information in a volume of data, and these modeling and viewing techniques have made possible new areas of application in diagnosis, surgical planning, and education. However, a variety of problems may cause visual illusions or artifacts in surface displays, and interpretation of volume rendering may require training and experience. The basic objective of three dimensional reconstructions is more effective communication of information in the image data, and thus it is important to understand any problems which can create misleading renderings since