Floating volume-of-interest in multilayer volume ray casting

摘要

A method that includes receiving vertices of a mesh, wherein the mesh represents a volume of interest (VOI) from an original volume, performing a transformation operation on the mesh, wherein the transformation operation transforms vertices of the mesh, obtaining sampling locations of the original volume using the transformed vertices, performing an inverse transformation on the transformed vertices of the mesh and the sampling locations, and performing a ray casting on the original volume to produce an image, wherein the ray casting uses the transformed vertices and a result of the inverse transformation as input.

主权项

1. A method, comprising:
receiving vertices of a mesh, wherein the mesh represents a volume of interest (VOI) from an original volume; performing a transformation operation on the mesh, wherein the transformation operation transforms vertices of the mesh, wherein the transformed vertices of the mesh can be anywhere in three-dimensional (3D) space so that a bounding box used in ray-layer buffer generation considers the transformed vertices; obtaining sampling locations of the original volume using the transformed vertices; performing an inverse transformation on the transformed vertices of the mesh and the sampling locations; rendering a cut-out of the VOI twice in each depth-peeling pass for creating a ray layer buffer, the first time as a static VOI and the second time as a floating VOI; and once the ray layer buffer is created, performing a ray casting on the original volume to produce an image, wherein the ray casting uses the transformed vertices and a result of the inverse transformation to define transformed starting positions of ray segments corresponding to regions inside the VOI.