Automated Radiotherapy Planning for Multi-Target Arc Beam Grouping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing radiation therapy planning methods struggle to efficiently treat multiple metastases (multimets) simultaneously with optimal dose distribution and minimal exposure to surrounding tissues, particularly when using limited beams, due to complex manual planning and challenges with multi-leaf collimator (MLC) openings.
Innovation Solution
An automated method for radiation therapy planning that optimally groups multiple targets into clusters, considers gantry and collimator angles, and adjusts MLC openings to achieve uniform dose distribution across targets, using conformal arc beams to minimize exposure to non-target areas.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If manual planning and grouping of multiple targets is performed, then treatment planner can consider various parameters, but planning time increases exponentially and optimal grouping cannot be guaranteed
Solution Approach 1:
The system enables automated treatment planning where the computer system performs target grouping and optimization autonomously without requiring manual intervention from the treatment planner, thus resolving the contradiction between planning quality and planning time
Solution Approach 2:
The patent replaces the manual mechanical process of treatment planning with an automated computer-based system that uses algorithms to evaluate target groupings, replacing human effort with computational power to achieve both speed and precision
2Productivity
If limited number of beams are used to treat multiple metastases simultaneously, then treatment time is reduced, but dose uniformity across targets becomes difficult to achieve
Solution Approach 1:
The system dynamically adjusts beam parameters such as collimator angle, couch angle, and gantry angle for each target group to optimize dose distribution, enabling dose uniformity to be maintained even when treating multiple targets with limited beams
Solution Approach 2:
The patent changes multiple parameters simultaneously (collimator angle, couch angle, gantry angle, MLC configuration) to achieve optimal dose uniformity across multiple targets while using a limited number of beams, resolving the contradiction between treatment throughput and dose precision
3Loss of time
If automated tools examine limited target groups and parameters, then computation time is reduced, but comprehensive optimization is not achieved
Solution Approach 1:
The system segments the automated planning process into manageable steps (target grouping, parameter optimization, dose calculation) while examining all possible target groups, achieving comprehensive automation without excessive computation time by processing information in organized segments
4Productivity
If MLC windows are adjusted to cover multiple targets, then treatment efficiency increases, but jagged openings occur leading to non-conformal dose delivery
Solution Approach 1:
The system dynamically optimizes MLC window configurations for each target group and beam angle, adjusting leaf positions to maintain conformal dose delivery while treating multiple targets efficiently, avoiding jagged openings through real-time parameter optimization
Data Source
Figure 1~2
Figure 3a~4b
Figure 5a~5b
AI summary
Disclosed herein is a treatment planning method for generating a treatment plan for radiation therapy where a set of targets (18a-18q) are to be treated, the method using a multi-leaf collimator (MLC) (4) for shaping an arc beam (24), a gantry (2) for holding the MLC, the gantry (2) capable of rotating at least partially around a patient, a couch (6) for positioning the patient, the MLC (4) being movable and defining a collimator angle (12), the gantry (2) being movable and defining a gantry angle (10) and the couch (6) optionally being movable thereby defining a couch angle (8), the method comprising the steps of: - Providing (S05) a number of candidate arc paths (20, 20', 20", 22) having an isocenter and a maximum number of arc beams (24) for the treatment plan and, depending on the maximum number of arc beams, a maximum number of target groups (34, 34', 34", 34‴, 34ʺʺ); - Providing (S06) a shape and position of each target (18a-18q) of the set of targets; - Calculating (S08) a target partition (36, 36', 36") out of at least some possible target partitions of the set of targets, the at least some possible target partitions (36, 36', 36") comprising a maximum number of target groups (34, 34', 34", 34‴, 34ʺʺ), each of the at least some possible target partitions (36, 36', 36") comprising target groups (34, 34', 34", 34‴, 34ʺʺ), - Determining (S09) a cost for each target group (34, 34', 34", 34‴, 34ʺʺ) of the possible target partition (36, 36', 36"), taking into account a candidate arc path (20, 20', 20", 22), at least one gantry angle (10) and at least one MLC angle (12) for each target group (34, 34', 34", 34‴, 34ʺʺ) of a possible target partition (36, 36', 36"); - Determining (S10) a cost for the possible target partition (36, 36', 36") and current candidate arc path (20, 20', 20", 22) by summing the costs of the target groups (34, 34', 34", 34‴, 34ʺʺ) of the possible target partition (36, 36', 36"); - Repeating (S12) the calculating (S08), determining (S09) and summing (S10) step for each of the at least some possible target partitions (36, 36', 36") and at least some of the candidate arc paths (20, 20', 20", 22), and - Selecting (S13) the optimal target partition (36, 36', 36") and candidate arc paths (20, 20', 20", 22) with the lowest sum of cost for the treatment plan.