Fluence map generation method, treatment plan generation method, and electronic device

Patent metadata

Field	Value
Publication number	US-12533529-B2
Application number	US-202318379527-A
Country	US
Kind code	B2
Filing date	Oct 12, 2023
Priority date	Jun 30, 2023
Publication date	Jan 27, 2026
Grant date	Jan 27, 2026

Abstract

A fluence map generation method includes: determining radiation information of a target volume at each beam angle of a plurality of beam angles, the target volume including a target area and organs at risk; and inputting the radiation information of the target volume at each beam angle into a fluence map generative model based on a denoising diffusion probabilistic model to obtain a fluence map corresponding to each beam angle. The fluence map corresponding to each beam angle is used to indicate an intensity of each beamlet corresponding to each beam angle in the target volume.

First claim

What is claimed is:

1 . A fluence map generation method, executed by an electronic device, comprising:
determining radiation information of a target volume at each beam angle of a plurality of beam angles, the target volume including a target area and organs at risk; and inputting the radiation information of the target volume at each beam angle into a fluence map generative model based on a denoising diffusion probabilistic model to obtain a fluence map corresponding to each beam angle; wherein the fluence map corresponding to each beam angle is used to indicate an intensity of each beamlet corresponding to each beam angle in the target volume.

2 . The fluence map generation method according to claim 1 , wherein inputting the radiation information of the target volume at each beam angle into the fluence map generative model based on the denoising diffusion probabilistic model to obtain the fluence map corresponding to each beam angle includes:
inputting a dose distribution of the target volume at each beam angle into the fluence map generative model based on the denoising diffusion probabilistic model to obtain the fluence map corresponding to each beam angle, wherein the dose distribution of the target volume at each beam angle is used to indicate a dose of radiation received by the target volume at each beam angle; or inputting an initial fluence map of the target volume at each beam angle into the fluence map generative model based on the denoising diffusion probabilistic model to obtain the fluence map corresponding to each beam angle, wherein the initial fluence map of the target volume at each beam angle of the plurality of beam angles is used to indicate an initial intensity of each beamlet corresponding to each beam angle in the target volume.

3 . The fluence map generation method according to claim 2 , wherein determining the radiation information of the target volume at each beam angle of the plurality of beam angles includes:
determining the dose distribution of the target volume at each beam angle of the plurality of beam angles, including:
obtaining an image and contour data of the target volume;
inputting the image and the contour data of the target volume into a field dose prediction model to obtain a predicted three-dimensional dose distribution of the target volume at each beam angle; and
determining the predicted three-dimensional dose distribution of the target volume at each beam angle as the dose distribution of the target volume at each beam angle.

4 . The fluence map generation method according to claim 3 , wherein determining the dose distribution of the target volume at each beam angle of the plurality of beam angles further includes:
projecting the predicted three-dimensional dose distribution of the target volume at each beam angle onto a projection plane orthogonal to each beam angle to obtain the dose distribution of the target volume at each beam angle.

5 . The fluence map generation method according to claim 2 , wherein determining the radiation information of the target volume at each beam angle of the plurality of beam angles includes:
determining the initial fluence map of the target volume at each beam angle of the plurality of beam angles, including:
obtaining an image and contour data of the target volume; and
inputting the image and the contour data of the target volume into a neural network model used for fluence map prediction, so as to obtain the initial fluence map of the target volume at each beam angle of the plurality of beam angles.

6 . The fluence map generation method according to claim 1 , wherein the fluence map generative model based on the denoising diffusion probabilistic model includes a first fluence map generative model or a second fluence map generative model;
in a case where the fluence map generative model based on the denoising diffusion probabilistic model includes the first fluence map generative model, inputting the radiation information of the target volume at each beam angle into the fluence map generative model based on the denoising diffusion probabilistic model to obtain the fluence map corresponding to each beam angle includes: inputting the radiation information of the target volume at each beam angle into the first fluence map generative model to obtain the fluence map corresponding to each beam angle; in a case where the fluence map generative model based on the denoising diffusion probabilistic model includes the second fluence map generative model, the fluence map generation method further comprises: obtaining structural data of the target volume at each beam angle of the plurality of beam angles; and inputting the radiation information of the target volume at each beam angle into the fluence map generative model based on the denoising diffusion probabilistic model to obtain the fluence map corresponding to each beam angle includes: inputting the structural data of the target volume at each beam angle and the radiation information into the second fluence map generative model to obtain the fluence map corresponding to each beam angle.

7 . A non-transitory computer-readable storage medium having stored instructions, wherein the instructions in the non-transitory computer-readable storage medium, when executed by an electronic device, cause the electronic device to execute the fluence map generation method according to claim 1 .

8 . A computer program product, comprising computer instructions carried on a non-transitory computer-readable storage medium, wherein the computer instructions, when run on an electronic device, cause the electronic device to perform the fluence map generation method according to claim 1 .

9 . A treatment plan generation method, executed by an electronic device, comprising:
obtaining an image and contour data of a target volume, where the target volume includes a target area and organs at risk; determining radiation information of the target volume at each beam angle of a plurality of beam angles according to the image and the contour data of the target volume; inputting the radiation information of the target volume at each beam angle into a fluence map generative model based on a denoising diffusion probabilistic model to obtain a fluence map corresponding to each beam angle, wherein the fluence map corresponding to each beam angle is used to indicate an intensity of each beamlet corresponding to each beam angle in the target volume; and generating a treatment plan according to the fluence map corresponding to each beam angle.

10 . The treatment plan generation method according to claim 9 , wherein inputting the radiation information of the target volume at each beam angle into the fluence map generative model based on the denoising diffusion probabilistic model to obtain the fluence map corresponding to each beam angle includes:
inputting a dose distribution of the target volume at each beam angle into the fluence map generative model based on the denoising diffusion probabilistic model to obtain the fluence map corresponding to each beam angle, wherein the dose distribution of the target volume at each beam angle is used to indicate a dose of radiation received by the target volume at each beam angle; or inputting an initial fluence map of the target volume at each beam angle into the fluence map generative model based on the denoising diffusion probabilistic model to obtain the fluence map corresponding to each beam angle, wherein the initial fluence map of the target volume at each beam angle of the plurality of beam angles is used to indicate an initial intensity of each beamlet corresponding to each beam angle in the target volume.

11 . The treatment pla

Assignees

• Our United Corp
• Shenzhen Our New Medical Tech Development Co Ltd

Inventors

• Guo Xiangyu
• Li Jinsheng

Classifications

• A61B6/5282

  due to scatter · CPC title

• A61N5/1039 Primary

  using functional images, e.g. PET or MRI · CPC title

• A61N5/1031

  using a specific method of dose optimization · CPC title

• G06T2207/20084

  Artificial neural networks [ANN] · CPC title

• G06T2207/20081

  Training; Learning · CPC title