Dynamic adaptation of radiotherapy treatment plans
A reference radiotherapy treatment plan having a plurality of control points, such as a sequence of gantry angles, aperture leaf positions, and intensity weights, can be adapted during the radiotherapy treatment session. Treatment imaging data of a patient may be obtained during the radiotherapy treatment session and one or more parameters may be determined using the treatment imaging data. Then, a current radiotherapy treatment plan may be generated based on the parameter(s). The reference radiotherapy treatment plan for the radiotherapy treatment session may be modified during the radiotherapy treatment session by updating one of the plurality of control points of the reference radiotherapy treatment plan with a control point of a current radiotherapy treatment plan, which may compensate for patient deformations during the delivery of radiotherapy treatment, or intrafraction patient deformations and thereby improve the delivery accuracy and efficacy of radiation doses to a patient undergoing radiotherapy treatment.
1 . A computer-implemented method for adaptation of a reference radiotherapy treatment plan having a plurality of control points, performed on a subject in a radiotherapy treatment session, the computer-implemented method comprising:
obtaining treatment imaging data of the subject during the radiotherapy treatment session;
determining at least one parameter using the treatment imaging data;
generating a current radiotherapy treatment plan based on the at least one parameter;
modifying the reference radiotherapy treatment plan for the radiotherapy treatment session during the radiotherapy treatment session by updating one of the plurality of control points of the reference radiotherapy treatment plan with a control point of the current radiotherapy treatment plan; and
using a trained model that includes:
generating at least two training images representing potential anatomical states of the subject, the at least two training images corresponding to at least two training parameters:
generating at least two training radiotherapy treatment plans corresponding to the at least two training images; and
generating a training regression between the at least two training parameters and the at least two training radiotherapy treatment plans.
2 . The computer-implemented method of claim 1 , comprising:
generating a current image of the subject using the at least one parameter and a reference image.
3 . The computer-implemented method of claim 2 , comprising:
generating a current structure set of the subject using the at least one parameter and a reference structure set.
4 . The computer-implemented method of claim 3 , comprising:
generating a current dose distribution using the current radiotherapy treatment plan and the current image.
5 . The computer-implemented method of claim 4 , comprising:
generating a dose volume histogram using the current dose distribution and the current structure set.
6 . The computer-implemented method of claim 1 ,
wherein generating the current radiotherapy treatment plan based on at least one parameter includes:
generating, using the training regression, the current radiotherapy treatment plan from the at least one parameter.
7 . The computer-implemented method of claim 1 , wherein generating the at least two training radiotherapy treatment plans corresponding to the at least two training images includes:
modifying the reference radiotherapy treatment plan to the at least two training images using segment-aperture morphing.
8 . The computer-implemented method of claim 1 , comprising:
generating at least two reference images of the subject;
selecting one of the at least two reference images as a primary reference image;
generating a deformation vector field between the primary reference image and each of the other reference images; and
determining at least one principal component of the deformation vector field.
9 . The computer-implemented method of claim 8 , wherein generating the at least two training images includes:
generating at least two deformation vector fields using the at least one principal component and at least two principal component weights; and
generating the at least two training images by deforming the primary reference image using the at least two deformation vector fields.
10 . The computer-implemented method of claim 8 , wherein determining the at least one parameter includes:
determining principal component weights that generate a deformation vector field that, when used to deform the primary reference image, are consistent with the treatment imaging data, wherein the principal component weights are assigned as the at least one parameter.
11 . The computer-implemented method of claim 8 , wherein generating the at least two reference images of the subject includes:
generating the at least two reference images from a 4D dataset acquired with the subject set up for treatment, and prior to beam-on.
12 . The computer-implemented method of claim 1 , wherein an individual one of the control points includes aperture information.
13 . The computer-implemented method of claim 1 , wherein an individual one of the control points includes multi-leaf collimator leaf position information.
14 . The computer-implemented method of claim 1 , wherein an individual one of the control points includes dose rate information.
15 . The computer-implemented method of claim 1 , wherein obtaining the treatment imaging data of the subject during the radiotherapy treatment session includes:
obtaining the treatment imaging data of the subject during the radiotherapy treatment session using at least one of a kV imaging data, 2D MR slices, surface camera imaging data, cone-beam CT (CBCT) imaging data, or MV imaging data.
16 . A radiotherapy system configured to perform the computer-implemented method of claim 1 .
17 . A tangible or non-tangible computer readable medium encoded with instructions that, when executed by a processor, cause the processor to perform the computer-implemented method of claim 1 .
18 . A radiotherapy system for adaptation of a reference radiotherapy treatment plan having a plurality of control points, performed on a subject in a radiotherapy treatment session, the radiotherapy system comprising:
an image acquisition device configured to acquire measurements of the subject during the radiotherapy treatment session;
a processor configured to:
perform, at a first rate, a first computational loop to generate at least one parameter using the acquired measurements, wherein the at least one parameter represents an anatomical state of the subject;
perform, at a second rate that is independent of the first rate, a second computational loop to generate a current radiotherapy treatment plan based on the at least one parameter, wherein the current radiotherapy treatment plan includes a plurality of control points; and
modify a reference radiotherapy treatment plan for the radiotherapy treatment session during the radiotherapy treatment session by updating a control point of the reference radiotherapy treatment plan with the control point of the current radiotherapy treatment plan; and
a radiotherapy device configured to deliver a dose of radiation to an anatomical region of interest using the current radiotherapy treatment plan.
19 . The radiotherapy system of claim 18 , wherein the processor is configured to:
repeatedly perform the first computational loop; and
repeatedly perform the second computational loop.
20 . The radiotherapy system of claim 19 , comprising:
a memory storage device configured to store sequentially generated parameters while the processor is repeatedly performing the first computational loop,
wherein the processor is configured to perform the second computational loop based on sequentially generated parameters.
21 . The radiotherapy system of claim 20 , wherein the processor is further configured to:
synchronize at least one of the control points of the current radiotherapy treatment plan with the acquired measurements.
22 . The radiotherapy system of claim 18 , wherein the processor is configured to:
perform, at a third rate that is independent of the first or second rates, a third computational loop to generate a current image of the subject using the most recent parameter.
23 . The radiotherapy system of claim 18 , wherein the processor is configured to:
perform, at a fourth rate that is independent of the first, second, or third rates, a fourth computational loop to generate a current dose distribution using the modified reference therapy plan with the updated control point and the current image.
24 . The radiotherapy system of claim 18 , wherein the processor is configured to:
perform, at a fifth rate that is independent of the first, second, third, or fourth rates, a fifth computational loop to generate a current structure set of the subject using the most recent parameter.
25 . The radiotherapy system of claim 18 , wherein the control point includes a dose per beam.
26 . The radiotherapy system of claim 18 , wherein the control point includes a position of a leaf of a multi-leaf collimator.
27 . The radiotherapy system of claim 18 , wherein the control point includes information defining a point along an aperture.