Radiotherapy method, computer program and computer system
Inventors
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-9786093-B2
Publication Date
2017-10-10
Expiration Date
Abstract
A method of obtaining a 3D image of a part of a patient's body is disclosed, based on a fraction image having a limited field-of-view and complementing this with information from a planning image having a greater field-of-view. In the area outside of the fraction image field-of-view, contour and anatomical data from the planning image are used to complement the fraction image, by means of a contour-guided deformable registration between the planning image and the fraction image.
Core Innovation
The invention concerns obtaining a 3D composite image of at least a first portion of a patient’s body by combining information from a 3D planning image and a 3D fraction image. The planning image has a first field-of-view and the fraction image has a second field-of-view that is limited in at least one dimension compared to the first field-of-view. A model outline is created that includes at least one area of the patient’s body included in the first field-of-view but not in the second field-of-view.
The method appends the model outline to the fraction image to include at least one outlined portion in the fraction image outside the second field-of-view, thereby obtaining an intermediate image. A deformable registration is then performed between the outline of the planning image and the outline of the fraction image, and the result of the deformable registration is used to include material information from the area in the planning image inside of the outlined portion in the intermediate image.
The disclosed approach is directed to enhancing dose computation accuracy by avoiding estimating outside-FOV regions and instead mapping material information from the planning image into an enhanced composite image. The outline can be based on a statistical shape model, and the outline adaptation can use a cost map based on a distance transform, optionally after rigid initial overlap.
Claims Coverage
The independent claim is clm-00001. It covers a pipeline with 4 inventive features that constructs an intermediate image by appending an outline to a limited-field fraction image and then uses deformable registration to transfer material information from a planning image into a composite image. Dependent claims refine the outline representation, adaptation mechanism, registration initialization, imaging modality pairings, and the types of material information included, while also describing computer program product and computer system embodiments.
Obtaining planning and limited-field fraction images
Obtaining a 3D planning image corresponding to the first portion of the patient’s body and having a first field-of-view, and obtaining a 3D fraction image of the first portion of the patient’s body, the fraction image having a second field-of-view which is limited in at least one dimension compared to the first field-of-view.
Creating and appending a model outline for an intermediate image
Creating a model outline including at least one area of the patient’s body included in the first field-of-view but not in the second field-of-view, and appending the model outline to the fraction image, to include at least one outlined portion in the fraction image which is outside the second field-of-view to obtain an intermediate image.
Performing deformable registration between planning and fraction outlines
Performing a deformable registration between the outline of the planning image and the outline of the fraction image.
Using registration to include planning material information in the composite image
Using the result of the deformable registration to include material information from the area in the planning image inside of the outlined portion in the intermediate image, to obtain a composite image.
Using a statistical shape model as the model outline
Creating the model outline as a statistical shape model outline derived from the planning image.
Adapting the model outline using a distance-transform-based cost map
Adapting the model outline to the fraction image using a cost function based on a distance transform.
Rigid registration for initial overlap of the outline
Performing rigid registration between the outline of the planning image and the outline of the fraction image to determine initial overlap prior to adapting the model outline to the fraction image.
CT planning and CT or CBCT fraction images
The 3D planning image is a CT image and the fraction image is a CT or CBCT image.
Including material information such as density, atomic composition, and anatomical information
Including material information such as density information, atomic composition, and/or anatomical information.
Overall claim coverage centers on appending a planning-derived model outline to a limited-field fraction image to form an intermediate image, deformably registering planning and fraction outlines, and using the registration result to import material information from the planning image into a composite image. Dependent claims further specify statistical shape model outlines, distance-transform-based cost adaptation, optional rigid initial overlap, CT/CBCT modality combinations, and expanded categories of material information including density, atomic composition, and anatomical information.
Stated Advantages
Improves dose computation accuracy by mapping material information from the planning image instead of estimating outside-FOV regions.
Documented Applications
Radiotherapy 3D composite imaging by enhancing a limited field-of-view fraction image using contour and material information from a larger field-of-view planning image for dose computation accuracy.
Compatibility for imaging modalities including CT, CBCT, MRI, and synthetic CT in the composite imaging approach.
Interested in licensing this patent?