Simulation and interaction methods for intuitive

• Autores: Rosell Torres Moreno
• Directores de la Tesis: Miguel A. Otaduy Tristán (dir. tes.)
• Lectura: En la Universidad Rey Juan Carlos ( España ) en 2019
• Idioma: español
• Texto completo no disponible (Saber más ...)
• Resumen

  • In this thesis we have developed methods to endow medical volume images with interactive deformation capabilities.

    The methods take as input a volume image, e.g., an MRI or CT-scan, discretize the volume with tetrahedral meshes, and transform the deformation back to the volume image, effectively deforming the underlying anatomy.

    We have demostrated our results in a variety of anatomical examples and medical settings, and we have shown that our methods success in deforming both full organs as well as highresolution anatomical elements.

    Our novel methods stand on a set of fundamental scientific contributions. Specifically, we have developed a novel numerical coarsening method, and we have extended it to support high-resolution boundary conditions. In addition, we have designed methods for intuitive deformation of the medical volume images using a touchscreen. These methods allow a user to interact with the volume dataset directly with the fingers, while we map the motion of the fingers to a rigid transformation of 3D handles by solving efficiently a closed-form optimization problem.

    In this thesis we develop three novel deformation modalities to naturally deform a volume image: mouse picking, volumetric tool interaction, and direct finger manipulation on a touchscreen. These three modalities provide rich options to interact with the volume both using a tablet or a desktop computer, to inspect a patient’s dataset and plan an intervention.