Predictive 3D Dose Constraints for Radiation Treatment Planning

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing radiation treatment plans lack efficiency in discriminating between target volumes and adjacent tissues, often requiring laborious and time-consuming manual adjustments to achieve optimal therapeutic results while minimizing collateral effects.

Innovation Solution

A method and apparatus that utilize predicted three-dimensional radiation dose distribution information as an optimization constraint, incorporating clinical goals and normalization factors to guide an optimization algorithm, reducing the need for multiple planning attempts.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If manual adjustments are made to achieve optimal therapeutic results, then treatment precision is improved, but planning time increases significantly

Engineering Contradiction:

Improvetreatment precisionVSAvoidplanning time

Core Design Contradiction:

Manufacturing precisionVSLoss of time

Solution Approach 1:

The system performs preliminary dose prediction using machine learning models trained on historical treatment data before the actual treatment planning process. This preliminary action provides an initial optimized treatment plan that serves as a starting point, reducing the number of manual adjustment iterations needed and significantly cutting down planning time while maintaining treatment precision

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system implements feedback loops where the predicted dose distribution from the machine learning model is continuously compared with actual treatment outcomes and clinical goals. This feedback mechanism allows the optimization algorithm to automatically adjust treatment parameters, reducing reliance on manual adjustments and accelerating the planning process while preserving accuracy

Inventive Principle:

Principle #23Feedback

2Reliability

If multiple planning attempts are made to achieve desired results, then treatment optimization is improved, but labor intensity and error risk increase

Engineering Contradiction:

Improvetreatment optimizationVSAvoidlabor intensity

Core Design Contradiction:

ReliabilityVSEase of operation

Solution Approach 1:

The system enables self-service treatment planning through automated optimization algorithms that independently perform multiple planning attempts based on the predicted dose distribution. The machine learning model automatically iterates through optimization cycles, adjusting parameters and evaluating outcomes without requiring continuous manual intervention, thereby reducing labor intensity while maintaining high treatment optimization standards

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system replaces manual mechanical adjustment processes with automated computational optimization. The machine learning-based prediction model and optimization algorithm substitute for the manual trial-and-error process, automatically performing multiple planning attempts with higher precision and lower error risk, thus improving treatment optimization while reducing labor intensity

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentEP4635563A1Method and apparatus to facilitate optimizing a radiation treatment plan

Publication Date: 2025.10.22 SIEMENS HEALTHINEERS INTERNATIONAL AG

Figure 1
Figure 2

AI summary

A control circuit can access 202 predicted three-dimensional radiation dose distribution information and optimize 204 a radiation treatment plan as a function, at least in part, of the predicted three-dimensional radiation dose distribution information to thereby prompt optimization towards the predicted three-dimensional radiation dose distribution information. By one approach, these teachings will support using the predicted three-dimensional radiation dose distribution information as an optimization constraint.