Adapting Movement Models via Real-Time Image Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing medical imaging technologies are insufficient in providing accurate movement models of patients during treatment, as the models generated prior to treatment may become less accurate due to patient movement, positioning changes, or cancer progression, leading to inefficiencies in radiation therapy delivery.
Innovation Solution
A system and method for adapting a previously generated movement model in real-time by comparing images captured during treatment to a reference image, using deformation fields to adjust displacement vectors and generate a model adaptation field, which refines the movement model to reflect actual patient positions and movements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If a movement model is generated prior to treatment, then the treatment plan can be prepared in advance, but the model accuracy deteriorates over time due to patient movement, positioning changes, or cancer progression
Solution Approach 1:
A movement model is generated in advance from a series of images captured over a period of time before treatment, allowing the treatment plan to be prepared ahead of time while capturing the patient's movement patterns during that preliminary period
Solution Approach 2:
The system continuously updates the pre-generated movement model by comparing newly captured images during treatment against the reference image and model, using the calculated model adaptation field to refine and maintain accuracy throughout the treatment process
2Measurement precision
If images are captured during treatment, then real-time movement data is obtained, but the number and quality of images are insufficient to calculate a complete movement model
Solution Approach 1:
A complete movement model is generated in advance from a comprehensive series of images captured over a period of time before treatment, establishing a robust reference model that can then be adapted using fewer real-time images
Solution Approach 2:
Images captured during treatment are used to update and adapt the pre-generated movement model through calculation of a model adaptation field, rather than attempting to generate a complete model from limited real-time images alone
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention provides a system and method for adapting a movement model based on an image captured during radiation of an object. The movement model is typically based on a series of images of an object captured over a period of time. The movement model and/or one or images included therein may be used to generate a reference image of the object. This reference image is compared with an image of the object optionally captured during object observation. The result of this comparison is used to adapt the movement model to conditions during the observation.