Radiotherapy Plan Generation Using Machine Learning

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current radiation therapy treatment planning is time-consuming and complex, often requiring a trial-and-error approach due to the need for precise control of radiation beams to minimize damage to healthy tissues while effectively targeting tumors, and lacks effective utilization of previous treatment plans to predict optimal parameters for new plans.

Innovation Solution

A computer-implemented method using machine learning techniques to generate radiotherapy treatment plans by processing input parameters, trained to establish relationships between input data and realizable treatment plan parameters, optimizing dose distribution and considering the physical constraints of the radiotherapy device, thereby reducing computational resources and improving plan quality.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If traditional optimization techniques are used to create treatment plans with multiple OARs, then the quality of dose distribution to tumor and critical organs is improved, but the time required to generate the treatment plan increases significantly

Engineering Contradiction:

Improvedose distribution qualityVSAvoidtreatment plan generation time

Core Design Contradiction:

Manufacturing precisionVSLoss of time

Solution Approach 1:

The system performs preliminary segmentation of the patient anatomy into target volume and organ-at-risk structures before optimization. By pre-processing the anatomical data and establishing initial dose constraints based on historical treatment plans, the system reduces the computational burden during the actual plan generation, thereby improving dose distribution quality without proportionally increasing planning time

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system utilizes historical treatment plans as templates or copies to guide the creation of new treatment plans. By copying proven dose distribution patterns and optimization parameters from previous successful treatments, the system can generate high-quality plans more efficiently, reducing both time and computational resources while maintaining dosimetric quality

Inventive Principle:

Principle #26Copying

2Reliability

If the number of organ-at-risk structures segmented is increased to improve coverage, then the completeness of anatomical modeling is improved, but the complexity of the treatment planning process increases

Engineering Contradiction:

Improveanatomical modeling completenessVSAvoidtreatment planning process complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system automatically segments the patient's anatomy into distinct target volumes and organ-at-risk structures using image processing algorithms. By dividing the complex anatomical space into manageable segmented regions, the system achieves comprehensive anatomical modeling while reducing the manual planning complexity, as the segmentation is performed automatically rather than requiring manual delineation of every structure

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system introduces an intermediary computational layer that automatically processes and organizes the segmented anatomical structures. This intermediary processing step manages the complexity by automatically handling the relationships between multiple OARs, dose constraints, and target volumes, thereby maintaining anatomical completeness without proportionally increasing planning complexity

Inventive Principle:

Principle #24Intermediary (Mediator)

3Manufacturing precision

If manual trial-and-error optimization is used to balance dose to tumor and sparing of OARs, then the clinical acceptability of the treatment plan is improved, but the computational resources and time required increase

Engineering Contradiction:

Improvedose optimization qualityVSAvoidplan generation efficiency

Core Design Contradiction:

Manufacturing precisionVSProductivity

Solution Approach 1:

The system implements feedback mechanisms by evaluating dose distribution metrics against clinical constraints and historical outcomes during the optimization process. By continuously monitoring dose to target and OARs and adjusting parameters based on this feedback, the system achieves clinically acceptable plans through automated iterative optimization, reducing the need for manual trial-and-error while maintaining dosimetric quality

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system automatically adjusts treatment plan parameters such as beam angles, intensities, and segmentation based on dose distribution feedback. By dynamically changing these parameters through automated optimization algorithms, the system achieves high-quality dose balancing between tumor coverage and OAR sparing without requiring extensive manual intervention, thereby improving both efficiency and productivity

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS11358003B2Generation of realizable radiotherapy plans

Publication Date: 2022.06.14 ELEKTA AB

AI summary

Techniques for generating a radiotherapy treatment plan are provided. The techniques include receiving an input parameter related to a patient, the input parameter being of a given type; processing the input parameter with a machine learning technique to estimate a realizable plan parameter of a radiotherapy treatment plan, wherein the machine learning technique is trained to establish a relationship between the given type of input parameter and a set of realizable radiotherapy treatment plan parameters to achieve a target radiotherapy dose distribution; and generating the radiotherapy treatment plan based on the estimated realizable plan parameter.