Granted Patent B2

US 9,375,184 · App. 14/467,246 · Granted Jun 28, 2016

Abstract

Claims (108)

1. A method for predicting respiratory motion of a patient, comprising:

generating a patient-specific anatomical model of the respiratory system from a 3D thoracic image of the patient;

deforming the patient-specific anatomical model of the respiratory system using a biomechanical model;

personalizing the biomechanical model for the patient by estimating a patient-specific thoracic pressure force field to drive the biomechanical model; and

predicting respiratory motion of the patient using the personalized biomechanical model driven by the patient-specific thoracic pressure force field,

wherein the patient-specific anatomical model includes at least one lung region, a thorax region, and a sub-diaphragm region, and generating the patient-specific anatomical model of the respiratory system from the 3D thoracic image of the patient comprises:

segmenting the at least one lung region, the thorax region, and the sub-diaphragm region in the 3D thoracic image;

generating a respective mesh for each of the segmented at least one lung region, thorax region, and sub-diaphragm region; and

defining a plurality of pressure zones on the meshes for the thorax region and the sub-diaphragm region.

2. The method of claim 1 , further comprising:

displaying a 3D+time respiratory motion map showing the predicted respiratory motion.

3. The method of claim 2 , further comprising:

performing one of image reconstruction or therapy delivery using the 3D+time respiratory motion map.

4. The method of claim 1 , wherein defining a plurality of pressure zones on the meshes for the thorax region and the sub-diaphragm region comprises:

defining pressure zones for the mesh of the thorax region by sub-dividing an inner surface of the thorax into a plurality of evenly spaced rings and splitting each ring into a plurality of pressure zones; and

defining pressure zones for the mesh of the sub-diaphragm region by splitting a superior surface of the sub-diaphragm region into a plurality of zones.

5. The method of claim 1 , wherein deforming the patient-specific anatomical model of the respiratory system using a biomechanical model comprises:

deforming the meshes for the at least one lung region, thorax region, and sub-diaphragm region based on forces due to pressures applied to the plurality of pressure zones on the thorax mesh and the sub-diaphragm mesh and based on interaction forces between the thorax mesh, the sub-diaphragm mesh and the at least one lung mesh calculated using a collision model that models sliding interactions between the meshes.

6. The method of claim 5 , wherein personalizing the biomechanical model for the patient by estimating a patient-specific thoracic pressure force field to drive the biomechanical model comprises:

estimating pressures applied to the pressure zones of the thorax mesh and the sub-diaphragm mesh to deform the lung mesh from two images corresponding to a first respiratory phase and a second respiratory phase.

7. The method of claim 6 , wherein estimating pressures applied to the pressure zones of the thorax mesh and the sub-diaphragm mesh to deform the lung mesh from two images corresponding to a first respiratory phase and a second respiratory phase comprises:

(a) performing an optimization algorithm to estimate optimal pressures applied to a single pressure zone on the thorax mesh and a single pressure zone one the sub-diaphragm mesh with predetermined initial values of the pressures;

(b) increasing a number of pressure zones on the thorax mesh and a number of pressure zones on the sub-diaphragm mesh;

(c) performing the optimization algorithm to estimate optimal pressures applied to the increased number of pressure zones on the thorax mesh and the sub-diaphragm mesh using previously estimated optimal pressure values as initial values; and

(d) repeating steps (b) and (c) until a target number of pressure zones on the thorax mesh and sub-diaphragm mesh is reached.

8. The method of claim 1 , wherein generating a patient-specific anatomical model of the respiratory system from a 3D thoracic image of the patient comprises:

generating a first patient-specific anatomical model from a first 3D thoracic image at a first respiratory phase; and

generating a second patient-specific anatomical model from a second 3D thoracic images at a second respiratory phase.

9. The method of claim 8 , wherein deforming the patient-specific anatomical model of the respiratory system using a biomechanical model comprises:

deforming the first patient-specific anatomical model from the first respiratory phase to the second respiratory phase using the biomechanical model.

10. The method of claim 9 , wherein personalizing the biomechanical model for the patient by estimating a patient-specific thoracic pressure force field to drive the biomechanical model comprises:

comparing the deformed first patient-specific anatomical model with the second patient-specific anatomical model; and

estimating a thoracic pressure force field that minimizes a cost function based on the comparison of the deformed first patient-specific anatomical model with the second patient-specific anatomical model.

11. The method of claim 10 , wherein the first respiratory phase is an end-exhale (EE) phase and the second respiratory phase is an end-inhale (EI) phase.

12. The method of claim 1 , wherein predicting respiratory motion of the patient using the personalized biomechanical model driven by the patient-specific thoracic pressure force field comprises:

modulating the patient-specific thoracic pressure force field; and

deforming the patient-specific anatomical model using the personalized biomechanical model driven by the modulated patient-specific thoracic pressure force field.

13. The method of claim 1 , wherein personalizing the biomechanical model for the patient by estimating a patient-specific thoracic pressure force field to drive the biomechanical model comprises:

estimating the patient-specific pressure force field to drive the model from a plurality of images corresponding to a plurality of respiratory phases.

14. An apparatus for predicting respiratory motion of a patient, comprising:

means for generating a patient-specific anatomical model of the respiratory system from a 3D thoracic image of the patient;

means for deforming the patient-specific anatomical model of the respiratory system using a biomechanical model;

personalizing the biomechanical model for the patient by estimating a patient-specific thoracic pressure force field to drive the biomechanical model; and

means for predicting respiratory motion of the patient using the personalized biomechanical model driven by the patient-specific thoracic pressure force field,

wherein the patient-specific anatomical model includes at least one lung region, a thorax region, and a sub-diaphragm region, and the means for generating a patient-specific anatomical model of the respiratory system from a 3D thoracic image of the patient comprises:

means for segmenting the at least one lung region, the thorax region, and the sub-diaphragm region in the 3D thoracic image;

means for generating a respective mesh for each of the segmented at least one lung region, thorax region, and sub-diaphragm region; and

means for defining a plurality of pressure zones on the meshes for the thorax region and the sub-diaphragm region.

15. The apparatus of claim 14 , further comprising:

means for displaying a 3D+time respiratory motion map showing the predicted respiratory motion.

16. The apparatus of claim 15 , further comprising:

means for performing one of image reconstruction or therapy delivery using the 3D+time respiratory motion map.

17. The apparatus of claim 14 , wherein the means for deforming the patient-specific anatomical model of the respiratory system using a biomechanical model comprises:

means for deforming the meshes for the at least one lung region, thorax region, and sub-diaphragm region based on forces due to pressures applied to the plurality of pressure zones on the thorax mesh and the sub-diaphragm mesh and based on interaction forces between the thorax mesh, the sub-diaphragm mesh and the at least one lung mesh calculated using a collision model that models sliding interactions between the meshes.

18. The apparatus of claim 17 , wherein the means for personalizing the biomechanical model for the patient by estimating a patient-specific thoracic pressure force field to drive the biomechanical model comprises:

means for estimating pressures applied to the pressure zones of the thorax mesh and the sub-diaphragm mesh to deform the lung mesh from two images corresponding to a first respiratory phase and a second respiratory phase.

19. The apparatus of claim 14 , wherein the means for generating a patient-specific anatomical model of the respiratory system from a 3D thoracic image of the patient comprises:

means for generating a first patient-specific anatomical model from a first 3D thoracic image at a first respiratory phase; and

means for generating a second patient-specific anatomical model from a second 3D thoracic images at a second respiratory phase.

20. The apparatus of claim 19 , wherein the means for deforming the patient-specific anatomical model of the respiratory system using a biomechanical model comprises:

means for deforming the first patient-specific anatomical model from the first respiratory phase to the second respiratory phase using the biomechanical model.

21. The apparatus of claim 20 , wherein the means for personalizing the biomechanical model for the patient by estimating a patient-specific thoracic pressure force field to drive the biomechanical model comprises:

means for comparing the deformed first patient-specific anatomical model with the second patient-specific anatomical model; and

means for estimating a thoracic pressure force field that minimizes a cost function based on the comparison of the deformed first patient-specific anatomical model with the second patient-specific anatomical model.

22. The apparatus of claim 21 , wherein the first respiratory phase is an end-exhale (EE) phase and the second respiratory phase is an end-inhale (EI) phase.

23. The apparatus of claim 14 , wherein the means for predicting respiratory motion of the patient using the personalized biomechanical model driven by the patient-specific thoracic pressure force field comprises:

means for modulating the patient-specific thoracic pressure force field; and

means for deforming the patient-specific anatomical model using the personalized biomechanical model driven by the modulated patient-specific thoracic pressure force field.

24. The apparatus of claim 14 , wherein the means for personalizing the biomechanical model for the patient by estimating a patient-specific thoracic pressure force field to drive the biomechanical model comprises:

means for estimating the patient-specific pressure force field to drive the model from a plurality of images corresponding to a plurality of respiratory phases.

25. A non-transitory computer readable medium storing computer program instructions for predicting respiratory motion of a patient, the computer program instructions when executed by a processor cause the processor to perform operations comprising:

generating a patient-specific anatomical model of the respiratory system from a 3D thoracic image of the patient;

deforming the patient-specific anatomical model of the respiratory system using a biomechanical model;

personalizing the biomechanical model for the patient by estimating a patient-specific thoracic pressure force field to drive the biomechanical model; and

predicting respiratory motion of the patient using the personalized biomechanical model driven by the patient-specific thoracic pressure force field, wherein the patient-specific anatomical model includes at least one lung region, a thorax region, and a sub-diaphragm region, and generating the patient-specific anatomical model of the respiratory system from the 3D thoracic image of the patient comprises:

segmenting the at least one lung region, the thorax region, and the sub-diaphragm region in the 3D thoracic image;

generating a respective mesh for each of the segmented at least one lung region, thorax region, and sub-diaphragm region; and

defining a plurality of pressure zones on the meshes for the thorax region and the sub-diaphragm region.

26. The non-transitory computer readable medium of claim 25 , wherein the operations further comprise:

displaying a 3D+time respiratory motion map showing the predicted respiratory motion.

27. The non-transitory computer readable medium of claim 26 , wherein the operations further comprise:

performing one of image reconstruction or therapy delivery using the 3D+time respiratory motion map.

28. The non-transitory computer readable medium of claim 25 , wherein defining a plurality of pressure zones on the meshes for the thorax region and the sub-diaphragm region comprises:

defining pressure zones for the mesh of the thorax region by sub-dividing an inner surface of the thorax into a plurality of evenly spaced rings and splitting each ring into a plurality of pressure zones; and

defining pressure zones for the mesh of the sub-diaphragm region by splitting a superior surface of the sub-diaphragm region into a plurality of zones.

29. The non-transitory computer readable medium of claim 25 , wherein deforming the patient-specific anatomical model of the respiratory system using a biomechanical model comprises:

deforming the meshes for the at least one lung region, thorax region, and sub-diaphragm region based on forces due to pressures applied to the plurality of pressure zones on the thorax mesh and the sub-diaphragm mesh and based on interaction forces between the thorax mesh, the sub-diaphragm mesh and the at least one lung mesh calculated using a collision model that models sliding interactions between the meshes.

30. The non-transitory computer readable medium of claim 29 , wherein personalizing the biomechanical model for the patient by estimating a patient-specific thoracic pressure force field to drive the biomechanical model comprises:

estimating pressures applied to the pressure zones of the thorax mesh and the sub-diaphragm mesh to deform the lung mesh from two images corresponding to a first respiratory phase and a second respiratory phase.

31. The non-transitory computer readable medium of claim 30 , wherein estimating pressures applied to the pressure zones of the thorax mesh and the sub-diaphragm mesh to deform the lung mesh from two images corresponding to a first respiratory phase and a second respiratory phase comprises:

(a) performing an optimization algorithm to estimate optimal pressures applied to a single pressure zone on the thorax mesh and a single pressure zone one the sub-diaphragm mesh with predetermined initial values of the pressures;

(b) increasing a number of pressure zones on the thorax mesh and a number of pressure zones on the sub-diaphragm mesh;

(c) performing the optimization algorithm to estimate optimal pressures applied to the increased number of pressure zones on the thorax mesh and the sub-diaphragm mesh using previously estimated optimal pressure values as initial values; and

(d) repeating steps (b) and (c) until a target number of pressure zones on the thorax mesh and sub-diaphragm mesh is reached.

32. The non-transitory computer readable medium of claim 25 , wherein generating a patient-specific anatomical model of the respiratory system from a 3D thoracic image of the patient comprises:

generating a first patient-specific anatomical model from a first 3D thoracic image at a first respiratory phase; and

generating a second patient-specific anatomical model from a second 3D thoracic images at a second respiratory phase.

33. The non-transitory computer readable medium of claim 32 , wherein deforming the patient-specific anatomical model of the respiratory system using a biomechanical model comprises:

deforming the first patient-specific anatomical model from the first respiratory phase to the second respiratory phase using the biomechanical model.

34. The non-transitory computer readable medium of claim 33 , wherein personalizing the biomechanical model for the patient by estimating a patient-specific thoracic pressure force field to drive the biomechanical model comprises:

comparing the deformed first patient-specific anatomical model with the second patient-specific anatomical model; and

estimating a thoracic pressure force field that minimizes a cost function based on the comparison of the deformed first patient-specific anatomical model with the second patient-specific anatomical model.

35. The non-transitory computer readable medium of claim 34 , wherein the first respiratory phase is an end-exhale (EE) phase and the second respiratory phase is an end-inhale (EI) phase.

36. The non-transitory computer readable medium of claim 25 , wherein predicting respiratory motion of the patient using the personalized biomechanical model driven by the patient-specific thoracic pressure force field comprises:

modulating the patient-specific thoracic pressure force field; and

deforming the patient-specific anatomical model using the personalized biomechanical model driven by the modulated patient-specific thoracic pressure force field.

37. The non-transitory computer readable medium of claim 25 , wherein personalizing the biomechanical model for the patient by estimating a patient-specific thoracic pressure force field to drive the biomechanical model comprises:

estimating the patient-specific pressure force field to drive the model from a plurality of images corresponding to a plurality of respiratory phases.

Assignments (7)