AI Dose Deposition Boosting for Monte Carlo-Level Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing radiation therapy dose calculation methods, particularly for small tumors, suffer from inaccuracies due to suboptimal modeling of radiation behavior, leading to difficulties in creating precise treatment plans that minimize damage to healthy tissue while effectively targeting cancerous tumors.
Innovation Solution
An artificial intelligence model, such as a neural network, is trained to enhance the accuracy of dose deposition calculations by inferring relationships between different dosing algorithms, allowing it to mimic the results of more computationally intensive methods like Monte Carlo simulations without the need for their full execution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If Monte Carlo simulations are used to calculate dose deposition, then measurement precision is improved, but productivity deteriorates due to excessive computational time and resources
Solution Approach 1:
The patent creates simplified copies of the complex Monte Carlo simulation model using machine learning algorithms. These ML models are trained on Monte Carlo data to replicate its dosimetric accuracy while executing much faster, effectively creating a lightweight copy that maintains the precision benefits without the computational burden.
Solution Approach 2:
The system performs preliminary work by pre-training machine learning models on extensive Monte Carlo simulation datasets. This preliminary training phase captures the complex radiation transport physics relationships, enabling rapid inference during actual treatment planning without requiring real-time Monte Carlo computations.
2Productivity
If deterministic methods are used to simulate radiation behavior, then productivity is improved with faster convergence, but measurement precision deteriorates compared to Monte Carlo simulations
Solution Approach 1:
The machine learning model is trained to copy the dosimetric accuracy of Monte Carlo simulations while maintaining the computational efficiency of deterministic methods. The ML model learns the complex particle transport physics from Monte Carlo data and applies these patterns rapidly during inference, combining the strengths of both approaches.
Solution Approach 2:
The system transforms the problem from solving complex partial differential equations with fine mesh refinement to using a trained ML model that has already captured the physics relationships. This parameter change in the computational approach allows maintaining accuracy while dramatically reducing computational resources and time.
3Productivity
If conventional dose calculation algorithms are used, then productivity is maintained with faster computation, but measurement precision deteriorates for small tumors requiring accurate dose predictions
Solution Approach 1:
The machine learning model replicates the high-accuracy dosimetric predictions of Monte Carlo simulations specifically for small tumor scenarios. By training on Monte Carlo data representing various tumor sizes and geometries, the ML model captures the complex dose distribution patterns that conventional algorithms miss, delivering accurate predictions at clinical speeds.
Data Source
AI summary
Embodiments described herein provide for training an artificial intelligence model to boost dose depositions. The artificial intelligence model receives medical images and a dose deposition determined according to a first dose deposition model. The artificial intelligence model modifies the received dose deposition determined according to the first dose deposition model such that the dose deposition simulates a dose deposition determined by a second dose deposition model.