Methods and systems for reconstructing a 3D anatomical structure undergoing non-rigid motion
There are described methods and systems for reconstructing a 3D anatomical structure undergoing non-rigid motion. The method comprises obtaining a 3D reference volume of the anatomical structure of the body, the reference volume corresponding to the anatomical structure at a reference phase of a respiratory cycle; acquiring 2D images of the anatomical structure at m prior times T in ={t−m, . . . , t−2, t−1}; estimating a set of deformations of the 3D reference volume at times n future T out ={t, t+1, . . . , t+n} from a previously learned probability distribution conditioned on partial observations and anatomical information; applying a spatial transformation to the 3D reference volume based on the set of deformations; and displaying the reference volume post-spatial transformation as a motion-compensated anatomical structure for each time step iϵT out .
1. A method for reconstructing of, and target tracking and beam positioning in, a 3D anatomical structure undergoing non-rigid motion, the method comprising:
obtaining a 3D reference volume of the anatomical structure of the body, the reference volume corresponding to the anatomical structure at a reference phase of a respiratory cycle;
acquiring 2D images of the anatomical structure at m prior times T in ={t−m, . . . , t−2, t−1};
estimating a set of deformations of the 3D reference volume at n future times T out ={t, t+1, . . . , t+n} from a previously learned probability distribution conditioned on partial observations and anatomical information;
applying a spatial transformation to the 3D reference volume based on the set of deformations;
displaying the reference volume post-spatial transformation as a motion-compensated anatomical structure for each time step iϵT out ;
tracking, based on the reference volume post-spatial transformation, a target displacing in the anatomical structure; and
positioning a treatment beam at an expected position of the target for each time step iϵT out .
2. The method of claim 1 , wherein the reference phase of the respiratory cycle is an exhale phase.
3. The method of claim 1 , wherein the probability distribution is learned from a conditional variational auto-encoder architecture.
4. The method of claim 1 , wherein the 2D images are magnetic resonance images.
5. The method of claim 1 , wherein the anatomical structure comprises a tumor target.
6. The method of claim 1 , wherein the anatomical structure is a liver or a lung.
7. The method of claim 1 , wherein the method is performed during radiotherapy treatments.
8. The method of claim 7 , wherein the radiotherapy is external beam radiotherapy.
9. The method of claim 7 , wherein the 2D images are acquired in real-time during the radiotherapy treatments.
10. A system for reconstructing of, and target tracking and beam positioning in, a 3D anatomical structure undergoing non-rigid motion, the system comprising:
at least one processor; and
a non-transitory computer-readable medium having stored thereon program instructions executable by the at lest one processor for:
obtaining a 3D reference volume of the anatomical structure of the body, the reference volume corresponding to the anatomical structure at a reference phase of a respiratory cycle;
acquiring 2D images of the anatomical structure at m prior times T in ={t−m, . . . , t−2, t−1};
estimating a set of deformations of the 3D reference volume at n future times T out ={t, t+1, . . . , t+n} from a previously learned probability distribution conditioned on partial observations and anatomical information;
applying a spatial transformation to the 3D reference volume based on the set of deformations; and
displaying the reference volume post-spatial transformation as a motion-compensated anatomical structure for each time step iϵT out ;
tracking, based on the reference volume post-spatial transformation, a target displacing in the anatomical structure; and
positioning a treatment beam at an expected position of the target for each time step iϵT out .
11. The system of claim 10 , wherein the reference phase of the respiratory cycle is an exhale phase.
12. The system of claim 10 , wherein the probability distribution is learned from a conditional variational auto-encoder architecture.
13. The system of claim 10 , wherein the 2D images are magnetic resonance images.
14. The system of claim 10 , wherein the anatomical structure comprises a tumor target.
15. The system of claim 10 , wherein the anatomical structure is a liver or a lung.
16. The system of claim 10 , wherein the 3D anatomical structure is reconstructed during radiotherapy treatments.
17. The system of claim 16 , wherein the radiotherapy is external beam radiotherapy.
18. The system of claim 16 , wherein the 2D images are acquired in real-time during the radiotherapy treatments.
19. An image-guided radiotherapy system comprising:
an image acquisition device for acquiring 2D images of the anatomical structure at times m prior T in ={t−m, . . . , t−2, t−1};
a volume reconstruction device in accordance with claim 10 ; and
a target tracking and beam positioning device for tracking a tumor target in the anatomical structure and positioning a radiation beam at an expected position of the tumor target for each time step iϵT out .
20. The image-guided radiotherapy system of claim 19 , wherein the system is a Magnetic Resonance Imaging Guided Linear Accelerator.