Automatic Plan Optimization for Changing Patient Anatomy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current radiation therapy planning systems face challenges in automating intensity-modulated radiation therapy (IMRT) and volumetric-modulated arc therapy (VMAT) optimization, particularly due to high-dimensional optimization problems and non-intuitive user interfaces, leading to tedious, inconsistent, and non-optimal treatment plans that may not adapt well to changing patient anatomy during treatment.
Innovation Solution
An automatic plan optimization system that uses deformable image registration to compute an optimal treatment plan by factoring in the dose delivered from previous fractions, allowing for dynamic planning adjustments based on anatomical changes, driven by clinical goals and priorities, and incorporating an auto-planning solution to generate improved treatment plans.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual optimization methods are used for IMRT/VMAT planning, then treatment plans can be created, but the process is time-consuming and tedious
Solution Approach 1:
The system enables automatic self-optimization of treatment plans through iterative algorithms that automatically adjust beam intensities and angles based on dose constraints and objectives, eliminating the need for manual intervention in the optimization process while maintaining plan quality
Solution Approach 2:
The system performs preliminary dose calculations and optimization iterations before final plan approval, using pre-defined constraints and objectives to generate optimized treatment plans automatically, reducing the time required for manual plan creation
2Manufacturing precision
If standard optimization algorithms are used, then treatment plans can be generated, but the plans may not be optimal due to difficulty in navigating high-dimensional optimization spaces
Solution Approach 1:
The optimization problem is segmented into multiple manageable iterations, where each iteration focuses on specific dose constraints and objectives. The complex high-dimensional optimization is broken down into sequential steps that are easier to control and evaluate
Solution Approach 2:
The system employs dynamic optimization where constraints and objectives are adjusted iteratively based on intermediate dose calculation results. This allows the optimization process to adaptively navigate the high-dimensional space by making incremental adjustments rather than attempting to solve the entire problem at once
3Reliability
If treatment plans are based on pre-treatment CT scans, then planning can be completed, but accuracy is compromised during treatment due to inter-fractional anatomical variations
Solution Approach 1:
The system incorporates feedback from accumulated dose information from previous fractions into the optimization process. By calculating the actual dose delivered and comparing it with the planned dose, the system adjusts subsequent treatment plans to account for anatomical changes and ensure accurate dose delivery
Solution Approach 2:
The system performs preliminary calculations of the accumulated dose from previous fractions before generating the next treatment plan. This allows the optimization to proactively account for anatomical changes and dose delivery variations, improving the accuracy of subsequent treatments
4Measurement precision
If accumulated dose from previous fractions is factored in, then treatment accuracy improves, but the optimization process becomes more complex
Solution Approach 1:
The complex optimization process is segmented into distinct modules: one for calculating accumulated dose from previous fractions, another for updating treatment objectives based on accumulated dose, and a third for generating optimized beam parameters. This modular approach manages complexity while maintaining precision
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach results in more intuitive, optimized, and consistent radiation therapy planning, reducing patient doses and improving treatment accuracy by adapting to changing anatomy, leading to more effective and efficient treatment plans.
Implementation Method 1
The proposed methodology achieves an optimal plan for future prescription(s) by automatically factoring in the dose which got delivered from the previous fractions. In the context of changing patient anatomy, the discussed methodology automatically computes an optimal plan on the patient's secondary image by factoring in the dose delivered with the initial plan on the patient's primary image.
Data Source
AI summary
A therapy planning system and method generate an optimal treatment plan accounting for changes in anatomy. Therapy is delivered to the subject according to a first auto-planned optimal treatment plan based on a first image of a subject. A second image of the subject is received after a period of time. The second image is registered with the first image to generate a deformation map accounting for physiological changes. The second image is segmented into regions of interest using the deformation map. A mapped delivered dose is computed for each region of interest using the dose delivery goals and the deformation map. The first treatment plan is merged with the segmented regions of the second image and the mapped delivered dose during optimization.