Efficient shape-based algorithms for modeling patient specific anatomy from 3D medical images: applications in virtual endoscopy and surgery

Virtual reality environments provide highly interactive, natural control of the visualization process, significantly enhancing the scientific value of the data produced by medical imaging systems. Due to the computational and real-time display update requirements of virtual reality interfaces, the complexity of organ and tissue surfaces which can be displayed is limited. In this paper, we present two new algorithms for the production of anatomic surface models containing a pre-specified number of polygons from patient- or subject-specific volumetric image data sets. The advantage of these algorithms is that they efficiently tile complex surfaces with a specified number of polygons selected to optimize the trade-off between surface detail and real-time display rates. Surface detail is preserved by extracting key shape features from the segmented objects, which adaptively constrains the model tilers. To illustrate the utility of these models, we present an overview of their application in computed endoscopy and surgery planning as developed in the Virtual Reality Assisted Surgery Program (VRASP) in the Biomedical Imaging Resource.