VMAT Optimization Framework Using Direct Aperture Generation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current VMAT optimization methods are inefficient due to their stochastic nature, lack of global optimization, and reliance on greedy heuristic approaches, leading to suboptimal and inconsistent dosimetry results, especially when dealing with complex beam configurations and multiple arcs.

Innovation Solution

A novel optimization framework that uses a single optimization function incorporating a level set function to regularize aperture shapes, solved using a proximal-class primal-dual algorithm, allowing for simultaneous optimization of all beam angles and direct aperture generation without progressive sampling.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If stochastic optimization methods (simulated annealing) are used for VMAT beam configuration, then the optimization can handle complex beam orientations and mechanical constraints, but the results are inconsistent and do not guarantee global optimality

Engineering Contradiction:

Improveability to handle complex beam configurationsVSAvoidconsistency and optimality of results

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The patent replaces stochastic optimization methods (simulated annealing) with a deterministic direct aperture optimization framework. This substitution eliminates the randomness and inconsistency inherent in stochastic approaches while maintaining the ability to handle complex VMAT beam configurations through direct mathematical optimization of aperture shapes and weights.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent segments the optimization problem into direct aperture shape optimization and weight optimization, solving them sequentially within a unified framework. This segmentation allows for more reliable and reproducible results by breaking down the complex stochastic optimization into manageable deterministic sub-problems.

Inventive Principle:

Principle #1Segmentation

2Productivity

If greedy heuristic methods with progressive beam insertion are used, then computational burden is reduced, but global optimization is lost and dosimetry quality deteriorates

Engineering Contradiction:

Improvecomputational efficiencyVSAvoiddosimetry quality

Core Design Contradiction:

ProductivityVSManufacturing precision

Solution Approach 1:

The patent performs preliminary action by directly optimizing aperture shapes for all beam angles simultaneously from the beginning, rather than progressively inserting beams one by one. This preliminary comprehensive optimization ensures global optimality is achieved while maintaining computational efficiency through the direct aperture framework.

Inventive Principle:

Principle #10Preliminary action

3Manufacturing precision

If multiple arcs are used to achieve desired dosimetry, then dose distribution quality improves, but treatment time and computational complexity increase

Engineering Contradiction:

Improvedose distribution qualityVSAvoidtreatment time

Core Design Contradiction:

Manufacturing precisionVSLoss of time

Solution Approach 1:

The patent creates a universal direct aperture optimization framework that can handle both single-arc and multi-arc VMAT configurations. This multi-functional approach allows the same optimization methodology to be applied regardless of the number of arcs, enabling efficient computation while achieving desired dosimetry quality through appropriate arc selection.

Inventive Principle:

Principle #6Universality (Multi-functionality)

4Adaptability or versatility

If simulated annealing with random sampling is used, then aperture configurations can be explored, but results depend on initial conditions and parameter ordering making reproduction difficult

Engineering Contradiction:

Improveexploration of aperture configurationsVSAvoidreproducibility of results

Core Design Contradiction:

Adaptability or versatilityVSStability of the object's composition

Solution Approach 1:

The patent replaces the stochastic simulated annealing mechanism with a deterministic direct aperture optimization system. This substitution eliminates dependence on initial conditions and random sampling, ensuring that the same input always produces the same optimal aperture configurations, thereby achieving full reproducibility while maintaining versatility.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS10857386B2System and method for volumetric modulated arc therapy

Publication Date: 2020.12.08 RGT UNIV OF CALIFORNIA

AI summary

A system and method are provided for generating and delivering a volumetric modulated arc therapy (“VMAT”) plan to a patient. In some aspects, the method includes receiving a representation of a patient comprising information related to target and non-target volumes of interest and generating an objective function based on the representation of the patient. The method also includes performing an iterative optimization process using the objective function to generate a VMAT plan and generating a report according to the VMAT plan.