Radiation Therapy Plan Optimization via Score Function
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Radiation therapy treatment plans exhibit significant variability due to planner skill, leading to inconsistent quality and increased risk for patients, as the ability to meet specified objectives does not depend on technological parameters or planner demographics.
Innovation Solution
A method and system that optimize treatment plans using a score function dependent on a quality function, which automates the optimization process by calculating and updating quality function values, applying scale factors, and adjusting smoothing parameters to reduce dependency on individual planner skills, thereby improving consistency and quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If manual treatment plan optimization is performed by treatment planners, then the treatment plan can be customized based on planner expertise, but the quality varies significantly depending on planner skill level
Solution Approach 1:
The system enables automated treatment plan optimization where the computational system performs the optimization task independently using algorithmic methods. The score function and quality function work together to automatically evaluate and improve treatment plans without human intervention, eliminating dependency on individual planner skills while maintaining customization through parameterized objectives.
Solution Approach 2:
The system transforms the optimization process from manual parameter adjustment to automated parameter optimization by changing the state from human-driven to algorithm-driven. The score function takes quality function parameters and automatically adjusts treatment plan parameters to optimize the score, replacing manual planner expertise with systematic parameter optimization.
2Reliability
If automated optimization is implemented using score function and quality function, then consistency and quality are improved, but the system complexity increases
Solution Approach 1:
The score function serves as an intermediary between the quality function (which defines treatment objectives) and the treatment plan optimization process. Rather than directly implementing complex optimization logic, the system uses the score function as a mediator that translates quality requirements into optimizable metrics, simplifying the overall system architecture.
Solution Approach 2:
The system replaces manual mechanical optimization processes (planner expertise and manual adjustment) with computational algorithms. The automated optimization using score and quality functions substitutes human cognitive processes with mathematical optimization, reducing the need for complex human expertise while maintaining system manageability through algorithmic approaches.
3Measurement precision
If multiple quality metrics are evaluated manually, then comprehensive treatment plan assessment is achieved, but time consumption increases
Solution Approach 1:
The automated optimization process enables continuous evaluation of multiple quality metrics simultaneously without interruption. The score function continuously assesses treatment plans against multiple quality functions, eliminating the sequential manual evaluation process and maintaining continuous optimization until convergence is achieved.
Solution Approach 2:
The system replaces manual evaluation mechanics with automated computational evaluation. Multiple quality metrics are assessed simultaneously by the computer system using the score function, eliminating the time-consuming sequential manual review process while maintaining comprehensive evaluation accuracy through parallel computational assessment.
Data Source
Figure 1
Figure 2
Figure 3A~3B
AI summary
It is provided a method for optimizing a treatment plan for use in radiation therapy. The method is performed in a treatment planning system and comprises the steps of: obtaining a first quality function of the treatment plan, the first quality function yielding an output value based on an input treatment plan; obtaining a first pair of input values and a slope indicator, the input values associating a value of the first quality function to a value of a score; constructing a score function that maps values of the first quality function to values of the score by fitting a curve to the first pair of input values and the slope indicator; and optimizing the treatment plan with respect to the score function, by varying the treatment plan such that the value of the score function is either improved or constrained to a feasible range of score values.