Method and device for determining a transformation between two images of an anatomical structure

1. An image transformation method, performed by a computer, for determining a transformation between two medical images of an anatomical structure (L), respectively taken at a first and second point in time, said method comprising the steps of:

a) acquiring first and second image datasets which represent the first and second image, respectively;

b) determining an image-based transformation (V) between the first image and the second image;

c) acquiring an elastic model of the anatomical structure (L) at the first point in time, wherein the elastic model comprises a plurality of nodes (N 1 -N 4 ) and the relationships between the nodes (N 1 -N 4 );

d) calculating the displacements of a subset of the nodes (N 1 ) in accordance with the image-based transformation;

e) calculating the displacements of the other nodes (N 2 -N 4 ) in accordance with the relationships between the nodes (N 1 -N 4 ) as represented by the elastic model while keeping the displacements of the nodes in the subset of nodes (N 1 ) constant; and

f) calculating the transformation from the displacement vectors which represent the displacements of the nodes (N 1 -N 4 ) as compared to their positions at the first point in time.

2. The method according to claim 1 , further comprising the step of aligning the elastic model with the first image, wherein this step is performed between steps c) and d).

3. The method according to claim 1 , wherein the subset of nodes (N 1 ) comprises the nodes which exhibit a displacement which is larger than a predetermined threshold.

4. The method according to claim 1 , wherein the subset of nodes (N 1 ) comprises 10% or less of the total number of nodes in the elastic model.

5. The method according to claim 1 , wherein the subset of nodes (N 1 ) comprises nodes on the surface of the anatomical structure (L) only.

6. The method according to claim 1 , wherein the nodes (N 1 -N 4 ) are mass points and the relationship between the nodes is represented by springs (S) between neighbouring nodes.

7. The method according to claim 1 , wherein the elastic model comprises a grid of equidistant nodes.

8. The method according to claim 1 , wherein the elastic model represents different kinds of tissue within the anatomical structure (L).

9. The method according to claim 1 , wherein the image-based transformation (V) is calculated by motion estimation.

10. The method according to claim 1 , wherein the image-based transformation (V) is a vector field.

11. The method according to claim 10 , wherein the displacement of a node (N 1 ) from the subset of nodes is calculated by interpolating the vectors of the vector field (V) for which the initial points surround the node (N 1 ).

12. The method according to claim 1 , wherein calculating the transformation from the displacement vectors involves calculating a displacement vector field from the displacement vectors by interpolation and using or in particular providing the displacement vector field as the transformation.

13. The method according to claim 1 , wherein nodes are selected for inclusion in the subset of nodes (N 1 ) on the basis of an atlas of the anatomical structure (L).

14. An image transformation system, comprising:

at least one computer having at least one processor and associated memory, the memory having computer instructions stored thereon, the instructions, when executed by the processor, causes the computer to:

acquire first and second images which represent an anatomical structure at a respective first point in time and a second point in time;

determine a transformation between the first image and the second image;

acquire a model of the anatomical structure at the first point in time, the model includes a plurality of nodes and the relationships between the nodes and represents different kinds of tissue within the anatomical structure;

calculate the displacements of a first subset of nodes of the plurality of nodes in accordance with the transformation;

calculate the displacements of the remaining nodes of the plurality of nodes in accordance with the relationships between the nodes as represented by the model while keeping the displacements of the nodes in the first subset of nodes constant;

determine the displacement of the plurality of nodes from the first point in time to the second point in time to create a plurality of displacement vectors; and

calculate the transformation from the displacement vectors.

15. The image transformation system of claim 14 wherein the memory has further instructions stored thereon, the instructions, when executed by the processor, causes the computer to calculate the transformation from the displacement vectors by providing the plurality of displacement vectors as the transformation.

16. The image transformation system of claim 14 wherein the memory has further instructions stored thereon, the instructions, when executed by the processor, causes the computer to calculate a displacement vector field from the plurality of displacement vectors by interpolation.

17. A computer implemented image transformation method for determining a transformation between a first medical image and a second medical image of an anatomical structure, respectively taken at a first point in time and second point in time, comprising:

acquiring the first and second image;

determining an image-based transformation between the first image and the second image;

acquiring a model of the anatomical structure at the first point in time, wherein the model comprises a plurality of nodes and relationships between the nodes, the model representing the contour of the anatomical structure;

calculating the displacements of a subset of the nodes in accordance with the image-based transformation;

calculating the displacements of the other nodes in accordance with the relationships between the nodes as represented by the model while keeping the displacements of the nodes in the subset of nodes constant;

calculating the transformation from the displacement vectors which represent the displacements of the nodes as compared to their positions at the first point in time.