High-speed inter-modality image registration via iterative feature matching
First Claim
1. A method of aligning a plurality of radiation therapy beams with a treatment target of a patient, comprising the steps of:
- a) preprocessing, comprising the steps of;
1) conducting a computed tomography (CT) scan; and
2) assembling an intermediate 3-D image from said scan; and
b) treatment, comprising the steps of;
1) acquiring a set of at least two real images of position of said treatment target of a patient;
2) using said at least two real images to produce a first feature vector;
3) generating a set of at least two digitally reconstructed images (DRRs) from said intermediate 3-D image;
4) using said set of at least two digitally reconstructed images to produce a second feature vector;
5) calculating a difference between said first feature vector and said second feature vector; and
6) if said difference does not fall below a minimum allowable value, employing an algorithm to adjust positions of said radiation therapy beams such that said radiation therapy beams are aligned with said treatment target of said patient.
1 Assignment
0 Petitions
Accused Products
Abstract
This invention describes a method for aligning radiation therapy beams with a treatment target of a patient. Diagnostic computer tomography scanning is used to map the treatment target in the patient'"'"'s anatomy and to aid physicians in deciding where to aim the radiation therapy beams. Digitally reconstructed radiographs are generated from an intermediate 3-D image produced from the CT scans. These DRRs are compared with x-ray images taken of the treatment target'"'"'s position taken at the time of treatment. A transformation equation is used to compare the DRRs and the x-ray images. A minimization equation is used to determine the similarity between the two sets of images. If the difference between the two sets fall below a determined minimum value, then the position of the treatment target is deemed correct and the radiation therapy begins. If the difference does not fall below the determined minimum value, then either the radiation beams or the treatment target is adjusted and the process is repeated. This procedure may be used many times in the course of a single treatment session, depending on the movement of the patient.
236 Citations
21 Claims
-
1. A method of aligning a plurality of radiation therapy beams with a treatment target of a patient, comprising the steps of:
-
a) preprocessing, comprising the steps of; 1) conducting a computed tomography (CT) scan; and 2) assembling an intermediate 3-D image from said scan; and b) treatment, comprising the steps of; 1) acquiring a set of at least two real images of position of said treatment target of a patient; 2) using said at least two real images to produce a first feature vector; 3) generating a set of at least two digitally reconstructed images (DRRs) from said intermediate 3-D image; 4) using said set of at least two digitally reconstructed images to produce a second feature vector; 5) calculating a difference between said first feature vector and said second feature vector; and 6) if said difference does not fall below a minimum allowable value, employing an algorithm to adjust positions of said radiation therapy beams such that said radiation therapy beams are aligned with said treatment target of said patient. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20)
-
-
21. A method of aligning radiation therapy beams with a treatment target of a patient, comprising the steps of:
-
a) acquiring a plurality of current two-dimensional images indicative of a current position of said treatment target relative to a means for generating said therapy beams; b) from a three-dimensional image comprising said treatment target, generating a corresponding plurality of reference two-dimensional images indicative of a reference position of said treatment target relative to said means for generating said therapy beams, wherein said treatment target is aligned with said means for generating said therapy beams in said reference position; c) comparing current feature vectors defined by said current two-dimensional images with corresponding reference feature vectors defined by said reference two-dimensional images; d) adjusting said current position if said current feature vectors do not match said reference feature vectors; and e) applying said therapy beams if said current feature vectors match said reference feature vectors.
-
Specification