Computation of a geometric parameter of a cardiac chamber from a cardiac tomography data set
First Claim
1. ) A method for extracting a geometric parameter from a cardiac tomography data set, said method comprising the steps of:
- providing a cardiac tomography data set including a plurality of voxels, each voxel including a respective tri-dimensional position vector and a respective intensity value;
providing a region of interest including a plurality of voxels selected from the cardiac tomography data set, the region of interest including at least in part voxels from the cardiac tomography data set representative of a cardiac chamber of interest, the cardiac chamber of interest having a surface of interest;
computing an approximative model of the surface of interest on a basis of the region of interest using a non-parametric method, the approximative model of the surface of interest including a plurality of surface coordinates tri-dimensional vectors;
computing a plurality of normal vectors, each normal vector being associated with a respective surface coordinate from the approximative model of the surface of interest;
computing a plurality of biopsy data sets, each biopsy data set including a plurality of data points, each biopsy data set being representative of a plurality of voxels oriented in a direction defined by a respective normal vector selected from the plurality of normal vectors and including the surface coordinate associated with the respective normal vector;
computing an exact model of at least a portion of the surface of interest on a basis of the plurality of biopsy data sets, the exact model of the portion of the surface of interest including a plurality of surface coordinates tri-dimensional vectors, the exact model of the portion of the surface of interest being computed by expanding locally the approximative model of the surface of interest in directions defined by the plurality of normal vectors; and
computing the geometric parameter on a basis of at least one surface coordinate from the plurality of surface coordinates included in the exact model of the portion of the surface of interest.
1 Assignment
0 Petitions
Accused Products
Abstract
A method for extracting a geometric parameter from a cardiac tomography data set. The method includes providing a cardiac tomography data set and a region of interest, computing an approximative model of a surface of interest, computing a plurality of normal vectors associated with a respective surface coordinate from the approximative model of the surface of interest, computing a plurality of biopsy data sets, each biopsy data set being oriented along a respective normal vector from the plurality of normal vectors, computing an exact model of the surface of interest from the plurality of biopsy data sets, and computing the geometric parameter on a basis of the exact model of the surface of interest.
-
Citations
27 Claims
-
1. ) A method for extracting a geometric parameter from a cardiac tomography data set, said method comprising the steps of:
-
providing a cardiac tomography data set including a plurality of voxels, each voxel including a respective tri-dimensional position vector and a respective intensity value;
providing a region of interest including a plurality of voxels selected from the cardiac tomography data set, the region of interest including at least in part voxels from the cardiac tomography data set representative of a cardiac chamber of interest, the cardiac chamber of interest having a surface of interest;
computing an approximative model of the surface of interest on a basis of the region of interest using a non-parametric method, the approximative model of the surface of interest including a plurality of surface coordinates tri-dimensional vectors;
computing a plurality of normal vectors, each normal vector being associated with a respective surface coordinate from the approximative model of the surface of interest;
computing a plurality of biopsy data sets, each biopsy data set including a plurality of data points, each biopsy data set being representative of a plurality of voxels oriented in a direction defined by a respective normal vector selected from the plurality of normal vectors and including the surface coordinate associated with the respective normal vector;
computing an exact model of at least a portion of the surface of interest on a basis of the plurality of biopsy data sets, the exact model of the portion of the surface of interest including a plurality of surface coordinates tri-dimensional vectors, the exact model of the portion of the surface of interest being computed by expanding locally the approximative model of the surface of interest in directions defined by the plurality of normal vectors; and
computing the geometric parameter on a basis of at least one surface coordinate from the plurality of surface coordinates included in the exact model of the portion of the surface of interest. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24)
-
-
25. ) A computer readable storage medium containing a program element for execution by a computing device for extracting a geometric parameter from a cardiac tomography data set, said program element comprising:
-
An input module operative for receiving;
i) a cardiac tomography data set including a plurality of voxels, each voxel including a respective tri-dimensional position vector and a respective intensity value; and
ii) a region of interest including a plurality of voxels selected from the cardiac tomography data set, the region of interest including at least in part voxels from the cardiac tomography data set representative of a cardiac chamber of interest, the cardiac chamber of interest having a surface of interest;
a preprocessing module operative for;
i) computing an approximative model of the surface of interest on a basis of the region of interest using a non-parametric method, the approximative model of the surface of interest including a plurality of surface coordinates tri-dimensional vectors;
ii) computing a plurality of normal vectors, each normal vector being associated with a respective surface coordinate from the approximative model of the surface of interest;
iii) computing a plurality of biopsy data sets, each biopsy data set including a plurality of data points, each biopsy data set being representative of a plurality of voxels oriented in a direction defined by a respective normal vector selected from the plurality of normal vectors and including the surface coordinate associated with the respective normal vector; and
iv) computing an exact model of the surface of interest on a basis of the plurality of biopsy data sets, the exact model of the surface of interest including a plurality of surface coordinates tri-dimensional vectors, the exact model of the surface of interest being computed by expanding locally the approximative model of the surface of interest in directions defined by the plurality of normal vectors; and
A geometry module operative for computing and outputting the geometric parameter on a basis of at least one surface coordinate from the plurality of surface coordinates included in the exact model of the surface of interest.
-
-
26. ) A cardiac tomograph, said cardiac tomograph comprising:
-
a tomography data acquisition system for acquiring a cardiac tomography data set including a plurality of voxels, each voxel including a respective tri-dimensional position vector and a respective intensity value;
a storage medium connected to said tomography data acquisition system; and
a computing device connected to said tomography data acquisition system and to said storage medium;
wherein said computer readable storage medium contains a program element for execution by said computing device, said program element being adapted to extract a geometric parameter from the cardiac tomography data set, said program element including;
i) An input module operative for receiving;
(1) the cardiac tomography data set; and
(2) a region of interest including a plurality of voxels selected from the cardiac tomography data set, the region of interest including at least in part voxels from the cardiac tomography data set representative of a cardiac chamber of interest, the cardiac chamber of interest having a surface of interest;
ii) a preprocessing module operative for;
(1) computing an approximative model of the surface of interest on a basis of the region of interest using a non-parametric method, the approximative model of the surface of interest including a plurality of surface coordinates tri-dimensional vectors;
(2) computing a plurality of normal vectors, each normal vector being associated with a respective surface coordinate from the approximative model of the surface of interest;
(3) computing a plurality of biopsy data sets, each biopsy data set including a plurality of data points, each biopsy data set being representative of a plurality of voxels oriented in a direction defined by a respective normal vector selected from the plurality of normal vectors and including the surface coordinate associated with the respective normal vector; and
(4) computing an exact model of the surface of interest on a basis of the plurality of biopsy data sets, the exact model of the surface of interest including a plurality of surface coordinates tri-dimensional vectors, the exact model of the surface of interest being computed by expanding locally the approximative model of the surface of interest in directions defined by the plurality of normal vectors; and
iii) A geometry module operative for computing and outputting the geometric parameter on a basis of at least one surface coordinate from the plurality of surface coordinates included in the exact model of the surface of interest.
-
-
27. ) A method for performing a segmentation of a multi-dimensional data set, said method comprising the steps of:
-
providing a multi-dimensional data set including a plurality of voxels, each voxel including a respective multi-dimensional position vector and a respective intensity value;
providing a region of interest including a plurality of voxels selected from the multi-dimensional data set, the region of interest defining a surface of interest;
computing an approximative model of the surface of interest on a basis of the region of interest using a non-parametric method, the approximative model of the surface of interest including a plurality of surface coordinates multi-dimensional vectors;
computing a plurality of normal vectors, each normal vector being associated with a respective surface coordinate from the approximative model of the surface of interest;
computing a plurality of biopsy data sets, each biopsy data set including a plurality of data points, each biopsy data set being representative of a plurality of voxels oriented in a direction defined by a respective normal vector selected from the plurality of normal vectors and including the surface coordinate associated with the respective normal vector, wherein computing a plurality of biopsy data sets includes computing a multiresolution laplacian data set from the multi-dimensional data set, each of the biopsy data set being representative of the variance of the multiresolution laplacian data set at a plurality of voxels oriented in a direction defined by a respective normal vector selected from the plurality of normal vectors and including the surface coordinate associated with the respective normal vector; and
computing an exact model of at least a portion of the surface of interest on a basis of the plurality of biopsy data sets, the exact model of the portion of the surface of interest including a plurality of surface coordinates multi-dimensional vectors, the exact model of the portion of the surface of interest being computed by expanding locally the approximative model of the surface of interest in directions defined by the plurality of normal vectors, the exact model of the portion of the surface of interest including a plurality of surface coordinates multi-dimensional vectors at which the variance of the multiresolution laplacian data set is substantially minimal.
-
Specification