Device and method for reconstruction and visualization of projection data

1. An apparatus for tomosynthesis reconstruction and visualization of projection data, comprising:

a computerized data processor comprising a raw data processing unit, a reconstruction unit, and an image data processing unit;

said data processor comprising an input port supplied with raw 2D projection data, for a plurality of 2D projection, obtained from a detector unit of an x-ray imaging apparatus;

said raw data processing unit being configured to load at least one module in said raw data processing unit with said raw 2D projection data as a 3D data block, said at least one module comprising data processing programs, including a 3D tomosynthesis image reconstruction program, and parameters stored in said reconstruction unit and said image data processing unit;

said module in said raw data processing unit being configured to process the raw 2D projection data in parallel in said 3D data block by simultaneously filtering all of said 2D projections in said 3D data block in parallel, to obtain processed and filtered data;

said at least one module also being configured to reconstruct a tomosynthesis image from the processed said 3D tomosynthesis image reconstruction program and filtered data, according to and said parameters stored in the reconstruction unit; and

said module also being configured to cause a 3D volume rendering of the reconstructed tomosynthesis image to be displayed, according to programs and parameters stored in said image generation unit, said 3D volume rendering being comprised of individual slices, and said module being configured to make data for displaying at least some of said slices available before completion of an entirety of the reconstruction of said tomosynthesis image.

2. An apparatus as claimed in claim 1 wherein said module is configured to implement a rotation of the raw projection data.

3. An apparatus as claimed in claim 1 wherein said module is a first module, and wherein said data processor comprises a second module loaded with programs from said image processing unit that cause projection data stored along an I-axis in the u-v-direction of a coordinate system to be assembled within arbitrary slice planes in the coordinate system to implement a computerized filtering process that is implemented n times, with said slice planes being subjected to said filtering in parallel.

4. An apparatus as claimed in claim 1 wherein said module is a first module and wherein said data processor comprises a second module loaded with programs from said image processing unit to process projection data respectively representing at least two slice planes extending in the u-v-direction of a coordinate system, said raw projection data being stored in n u-v-parallel planes in said coordinate system, and being subjected to computerized filtering in parallel n times.

5. An apparatus as claimed in claim 1 wherein said module is a first module and wherein said data processor comprises a second module loaded with programs from said image processing unit, wherein projection data are stored in n u-v-parallel planes of a coordinate system and are assembled within n v-i slice planes in said coordinate system in a filtering process implemented n times, with projection data for at least two v-i slice planes being filtered in parallel.

6. An apparatus as claimed in claim 1 comprising a back-projection module controlled by a back-projection unit in said image processing unit to implement a back-projection of complete u-v planes of filtered 2D projection data in a reconstruction volume using 2D or 3D texture mapping with a parallel accumulation of mapped data in processed regions of the reconstruction volume.

7. An apparatus as claimed in claim 1 comprising a filter module controlled by a filter unit for filtering raw projection data on a slice-by-slice basis by processing the data in parallel line-by-line during the acquisition of the raw projection data for at least one slice, said filtering being executed n times.

8. A method for reconstruction and visualization of projection data, comprising the steps of:

configuring a computerized data processor to include a raw data processing unit, a reconstruction unit, and an image data processing unit;

supplying said data processor with raw 2D projection data, for a plurality of 2D projections obtained from a detector unit of an x-ray imaging apparatus, to said processor;

in said raw data processing unit, loading at least one module in said raw data processing unit with said raw 2D projection data as a 3D data block, said at least one module comprising data processing programs, including 3D tomosynthesis image reconstruction program, and parameters stored in said reconstruction unit and said image data processing unit;

processing the raw 2D projection data in said module in said raw data processing unit in parallel in said 3D data block by simultaneously filtering all of said 2D projections in said 3D data block in parallel, to obtain processed and filtered data;

also in said at least one module, reconstructing a tomosynthesis image from the processed and filtered data, according to said 3D tomosynthesis reconstruction program and said parameters stored in the reconstruction unit; and

said module also causing a 3D volume rendering of the reconstructed tomosynthesis image to be displayed, according to programs and parameters stored in said image generation unit, said 3D volume rendering being comprised of individual slices, and said module making data for displaying at least some of said slices available before completion of an entirety of the reconstruction of said tomosynthesis image.

9. A method as claimed in claim 8 comprising in said module implementing a rotation of the raw projection data.

10. A method as claimed in claim 8 wherein said module is a first module, and comprising loading a second module in said data processor comprises with programs from said image processing unit that cause projection data stored along an I-axis in the u-v-direction of a coordinate system to be assembled within arbitrary slice planes in the coordinate system to implement a computerized filtering process that is implemented n times, with said slice planes being subjected to said filtering in parallel.

11. A method as claimed in claim 8 wherein said module is a first module and comprising loading a second module in said data processor with programs from said image processing unit to process projection data respectively representing at least two slice planes extending in the u-v-direction of a coordinate system, said raw projection data being stored in n u-v-parallel planes in said coordinate system, and being subjected to computerized filtering in parallel n times.

12. A method as claimed in claim 8 wherein said module is a first module and comprising loading a second module loaded in said data processor with programs from said image processing unit, wherein projection data are stored in n u-v-parallel planes of a coordinate system and are assembled within n v-i slice planes in said coordinate system in a filtering process implemented n times, with projection data for at least two v-i slice planes being filtered in parallel.

13. A method as claimed in claim 8 comprising, in a back-projection module controlled by a back-projection unit in said image processing unit implementing a back-projection of complete u-v planes of filtered 2D projection data in a reconstruction volume using 2D or 3D texture mapping with a parallel accumulation of mapped data in processed regions of the reconstruction volume.

14. A method as claimed in claim 8 comprising, in a filter module controlled by a filter unit, filtering raw projection data on a slice-by-slice basis by processing the data in parallel line-by-line during the acquisition of the raw projection data for at least one slice, said filtering being executed n times.