Method for distinguishing between four materials in tomographic records of a dual-energy CT system
First Claim
1. A method for distinguishing between four materials (M1, M2, M3, M4) in tomographic records of a dual-energy Computed Tomography (CT) system, the method comprising:
- scanning of an examination area of an examination object, including at least four different materials (M1, M2, M3, M4), whereinthe scanning is carried out using two different X-ray energy spectra from a multiplicity of projection directions, and absorption data is measured for each X-ray spectrum that is used,Hounsfield Units (HU) values of four materials (M1, M2, M3, M4) form a quadrilateral on an HU-value diagram of the X-ray spectra used for scanning, the two diagonals (M1-M4, M2-M3) of which quadrilateral intersect one another,the materials which form one diagonal are each mixed, andtwo materials which define a diagonal form a mixture with particle sizes in the same order of magnitude as the resolution of the CT system;
reconstructing at least one first tomographic image data record from the absorption data of the first X-ray spectrum;
reconstructing at least one second tomographic image data record from the absorption data of the second X-ray spectrum;
defining a size of at least one of a two-dimensional and three-dimensional viewing area around each voxel whose material content is to be distinguished, and carrying out subsequent statistical calculations in the viewing area;
imaging adjacent voxels from the viewing area on an HU value diagram, for each voxel whose material content is to be distinguished;
calculating distances (x1, x2) from the diagonals (M1-M4, M2-M3) for all the imaged voxels, and forming a mean squares of the distances (x12, x22); and
assuming, when the mean square distance (x12, x22) to one diagonal is less than to the other diagonal, that the composition of the voxel is composed of the materials (M1+M4, M2+M3) to whose diagonal the lower mean square distance (x12, x22) occurs.
2 Assignments
0 Petitions
Accused Products
Abstract
A method and a CT system are disclosed having a computation unit for distinguishing between four materials (M1, M2, M3, M4) in tomographic records of a dual-energy CT system, wherein the size of a two-dimensional or three-dimensional viewing area is defined around each voxel whose material content is to be distinguished. In an embodiment of the method, the adjacent voxels from the viewing area on an HU value diagram are imaged for each voxel (Vi) whose material content is to be distinguished, the distances from the diagonals are calculated for all the imaged voxels, and the mean squares of these distances are formed (x12, x22). If the mean square distance (x12, x22) to one diagonal is less than to the other diagonal, then the composition of the voxel is assumed to be composed of the materials to whose diagonal the lower mean square distance (x12, x22) occurs.
16 Citations
21 Claims
-
1. A method for distinguishing between four materials (M1, M2, M3, M4) in tomographic records of a dual-energy Computed Tomography (CT) system, the method comprising:
-
scanning of an examination area of an examination object, including at least four different materials (M1, M2, M3, M4), wherein the scanning is carried out using two different X-ray energy spectra from a multiplicity of projection directions, and absorption data is measured for each X-ray spectrum that is used, Hounsfield Units (HU) values of four materials (M1, M2, M3, M4) form a quadrilateral on an HU-value diagram of the X-ray spectra used for scanning, the two diagonals (M1-M4, M2-M3) of which quadrilateral intersect one another, the materials which form one diagonal are each mixed, and two materials which define a diagonal form a mixture with particle sizes in the same order of magnitude as the resolution of the CT system; reconstructing at least one first tomographic image data record from the absorption data of the first X-ray spectrum; reconstructing at least one second tomographic image data record from the absorption data of the second X-ray spectrum; defining a size of at least one of a two-dimensional and three-dimensional viewing area around each voxel whose material content is to be distinguished, and carrying out subsequent statistical calculations in the viewing area; imaging adjacent voxels from the viewing area on an HU value diagram, for each voxel whose material content is to be distinguished; calculating distances (x1, x2) from the diagonals (M1-M4, M2-M3) for all the imaged voxels, and forming a mean squares of the distances (x12, x22); and assuming, when the mean square distance (x12, x22) to one diagonal is less than to the other diagonal, that the composition of the voxel is composed of the materials (M1+M4, M2+M3) to whose diagonal the lower mean square distance (x12, x22) occurs. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21)
-
Specification