Adaptive Radiotherapy Dose Control Using Real-Time Anatomy Deviations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing radiotherapy techniques fail to adjust radiotherapy device parameters in real-time within a treatment fraction, leading to over- or under-delivery of radiation to tumors and organs at risk, without considering the total influence of radiation across the entire treatment fraction.
Innovation Solution
A machine learning model is used to estimate and adjust radiotherapy device parameters in real-time during a treatment fraction by comparing current patient anatomy with a reference anatomy, optimizing dose delivery to maintain prescribed doses and minimize exposure to organs at risk.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If real-time adjustment of radiotherapy device parameters is implemented, then dose delivery accuracy is improved, but device complexity increases
Solution Approach 1:
A machine learning model serves as an intermediary between patient anatomy data and radiotherapy device parameter adjustments. The model processes anatomical deviations and predicts optimal parameter adjustments, eliminating the need for complex real-time physics calculations and simplifying the control system while maintaining dosimetric accuracy.
Solution Approach 2:
Traditional mechanical control systems based on real-time physics simulations are replaced with a data-driven machine learning approach. This substitution reduces computational complexity and enables faster parameter adjustments during treatment delivery.
2Manufacturing precision
If real-time monitoring and adjustment of patient anatomy is performed, then treatment accuracy is improved, but treatment time increases
Solution Approach 1:
The machine learning model is trained offline on extensive datasets of patient anatomies and corresponding optimal treatment parameters. This preliminary training phase enables the model to make rapid predictions during actual treatment without requiring time-consuming real-time calculations, thus maintaining treatment accuracy while minimizing treatment time.
Solution Approach 2:
The system automatically monitors patient anatomy deviations and adjusts radiotherapy parameters without requiring manual intervention. The closed-loop control system continuously feeds back anatomical data to the machine learning model, which autonomously determines parameter adjustments, eliminating delays associated with manual re-planning.
3Measurement precision
If machine learning model training is performed with extensive patient data, then model accuracy is improved, but data processing time increases
Solution Approach 1:
The training process is divided into separate phases: offline training on extensive datasets to build model accuracy, and online inference during treatment delivery for rapid predictions. This segmentation allows extensive data processing to occur beforehand without impacting treatment time, as the trained model requires minimal processing during actual therapy delivery.
Data Source
AI summary
Techniques for adjusting radiotherapy treatment for a patient in real-time are provided. The techniques include obtaining a training patient anatomy at a first time within a training radiotherapy treatment fraction after a training radiotherapy treatment dose has been delivered by a radiotherapy device; computing a deviation between the training patient anatomy at the first time and reference training patient anatomy during the training radiotherapy treatment fraction, wherein the reference training patient anatomy indicates a prescribed training dose parameter to be delivered within the training radiotherapy treatment fraction; applying the computed deviation to a machine learning model to estimate one or more intra-fraction radiotherapy treatment parameters of a function that provides a radiotherapy device parameter adjustment based on the one or more intra-fraction radiotherapy treatment parameters; and training the machine learning model to establish a relationship between the computed deviation and the one or more intra-fraction radiotherapy treatment parameters.