Rendering anatomical structures with their nearby surrounding area
View Patent ↗System and methods of displaying anatomical structures and their surrounding area, are disclosed. For a viewing point the anatomical structures are rendered separate from their surrounding and saved. The surrounding area of the anatomical structure within a viewing frustum is extracted, interpolated and rendered. The rendered anatomical structures and calculated image of the surrounding are combined into a complete rendering of the anatomical structures with its nearby surrounding areas.
1. A method of processing medical image data for smooth rendering from a defined point of view of one or more anatomical structures with a nearby surrounding area from 3D image data, wherein the one or more anatomical structures are separated from the nearby surrounding area, comprising:
creating an image by rendering with a processor the one or more separated anatomical structures from the defined point of view and saving a depth information of the rendered anatomical structures in a first Z-buffer with non-visible pixels set to value 1 and saving the rendered image of the anatomical structures;
extracting a boundary between the rendered one or more anatomical structures and the nearby surrounding area using the depth information of the rendered one or more anatomical structures;
rendering the nearby surrounding area from the defined point of view, including the steps of extracting from the first Z-buffer the pixels that lie on a boundary between visible and non-visible areas, storing each boundary pixel's image coordinates in a 2D array, adding four points to the 2D array as defined by a visualization frustum; creating a 2D mesh as defined by the 2D array points; and projecting the 2D mesh to a 3D mesh by assigning to each point defining the 2D mesh its corresponding 3D coordinate, and combining the renderings of the one or more anatomical structures and the nearby surrounding area to be displayed on a display.
2. The method as claimed in claim 1 , wherein rendering the nearby surrounding area further includes interpolating points on the 3D mesh to determine pixels in the nearby surrounding area.
3. The method as claimed in claim 1 , wherein rendering the nearby surrounding area further includes interpolating coordinates of points on the 3D mesh and accessing the medical image data based on the coordinates to determine pixels in the nearby surrounding area.
4. The method as claimed in claim 1 , wherein the 2D mesh is created by applying Delaunay triangulation.
5. The method as claimed in claim 1 , wherein second near-plane depth information is created, comprising the steps:
rendering the 3D mesh and saving the corresponding near-plane depth information of the rendered mesh as a second Z-buffer with non-visible pixels set to value 1; and
setting the pixels that corresponds to visible pixels in the first Z-buffer to value 0.
6. The method as claimed in claim 5 , wherein the one or more anatomical structures are rendered with the nearby surrounding area from 3D image data comprising:
creating an image by rendering the 3D image data using the second Z-buffer as the rendering rays entry points into the data;
propagating the rays a specified depth distance into the data;
skipping the pixels with second Z-buffer values of one; and
assigning to each pixel with second Z-buffer values of zero the corresponding value from the saved rendered one or more anatomical structures.
7. The method as claimed in claim 1 , wherein far-plane depth information is used for determining the rendering rays exit points from the data.
8. A system of processing medical image data for smooth rendering from a defined point of view of one or more anatomical structures with a nearby surrounding area from 3D image data, wherein the one or more anatomical structures are separated from the nearby surrounding area, comprising:
a processor;
software operable on the processor for:
creating an image by rendering the one or more separated anatomical structures from the defined point of view and saving a depth information of the rendered one or more anatomical structures in a first Z-buffer with non-visible pixels set to value 1 and saving the rendered image of the one or more anatomical structures;
extracting a boundary between the rendered one or more anatomical structures and the nearby surrounding area using the depth information of the rendered one or more anatomical structures;
rendering the nearby surrounding area from the defined point of view, including extracting from the first Z-buffer the pixels that lie on a boundary between visible and non-visible areas, storing each boundary pixel's image coordinates in a 2D array, adding four points to the 2D array as defined by a visualization frustum; creating a 2D mesh as defined by the 2D array points; and projecting the 2D mesh to a 3D mesh by assigning to each point defining the 2D mesh its corresponding 3D coordinate, and
combining the renderings of the one or more anatomical structures and the nearby surrounding area.
9. The system as claimed in claim 8 , wherein rendering the nearby surrounding area further includes interpolating points on the 3D mesh to determine pixels in the nearby surrounding area.
10. The system as claimed in claim 8 , wherein rendering the nearby surrounding area further includes interpolating coordinates of points on the 3D mesh and accessing the medical image data based on the coordinates to determine pixels in the nearby surrounding area.
11. The system as claimed in claim 8 , wherein the 2D mesh is created by applying Delaunay triangulation.
12. The system as claimed in claim 8 , wherein second near-plane depth information is created comprising:
rendering the 3D mesh and saving the corresponding near-plane depth information of the rendered mesh as a second Z-buffer with non-visible pixels set to value 1; and
setting the pixels that corresponds to visible pixels in the first Z-buffer to value 0.
13. The system as claimed in claim 12 , wherein the one or more anatomical structures are rendered with the nearby surrounding area from 3D image data comprising:
creating an image by rendering the 3D image data using the second Z-buffer as the rendering rays entry points into the data;
propagating the rays a specified depth distance into the data;
skipping the pixels with second Z-buffer values of one; and
assigning to each pixel with second Z-buffer values of zero the corresponding value from the saved rendered one or more anatomical structures.
14. The system as claimed in claim 8 , wherein far-plane depth information is used for determining the rendering rays exit points from the data.
15. The system as claimed in claim 8 , wherein the processor is a graphics processing unit.