Shape-Based Retrieval for Radiation Treatment Planning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In intensity-modulated radiation therapy, existing methods face challenges in optimizing radiation delivery to tumor volumes while minimizing exposure to adjacent organs at risk, as they lack effective tools to compare and leverage geometric and dosimetric data from similar patient cases for personalized treatment planning.
Innovation Solution
A shape-based retrieval system that uses overlap volume histograms and dose volume histograms to match new patient data with prior cases, allowing for the determination of optimal multi-leaf collimator settings and intensity distribution by analyzing the geometric relationships and proximity of organs at risk to the tumor, facilitating the retrieval of effective treatment plans from similar patients.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If existing treatment planning methods are used, then treatment plans can be determined, but they cannot effectively leverage geometric and dosimetric data from similar patient cases
Solution Approach 1:
The patent creates shape signatures that are simplified representations (copies) of complex patient anatomy and treatment plan geometry. These shape signatures capture essential geometric characteristics without requiring full detailed models, enabling efficient comparison and retrieval of similar cases while preserving the necessary information for treatment planning decisions.
Solution Approach 2:
The patent replaces manual comparison methods with automated computational algorithms that calculate shape signatures and perform similarity matching. This substitution of mechanical/manual processes with algorithmic systems enables efficient processing of geometric and dosimetric data from multiple patient cases without proportionally increasing system complexity.
2Productivity
If manual treatment planning is performed, then treatment plans can be created, but the process is time-consuming and lacks efficiency
Solution Approach 1:
The patent performs preliminary calculations of shape signatures and geometric characteristics during the treatment planning process itself, rather than requiring separate post-processing steps. This preliminary action embeds the analysis within the existing workflow, improving productivity without adding significant time overhead by utilizing data already available during plan creation.
Solution Approach 2:
The system automatically retrieves and compares similar patient cases without requiring manual intervention to search for precedents. The shape-based retrieval system self-services the treatment planning process by autonomously identifying relevant historical data and presenting it to planners, thereby reducing both time loss and improving productivity simultaneously.
3Measurement precision
If comprehensive patient data analysis is performed, then accurate treatment plans can be determined, but the complexity of data processing increases
Solution Approach 1:
The patent extracts only the essential geometric and dosimetric features needed for treatment planning decisions, rather than processing all available patient data. By taking out and isolating the critical shape characteristics and dose distribution metrics, the system achieves high measurement precision while avoiding the complexity of comprehensive data processing.
Solution Approach 2:
The patent applies different levels of analysis to different aspects of treatment planning. Shape signatures focus specifically on geometric relationships between target volumes and organs at risk, while dosimetric analysis concentrates on dose distribution characteristics. This local quality approach ensures precise measurement of each aspect without requiring equally complex processing of all data simultaneously.
Data Source
AI summary
A system and method for determining at least one new treatment plan for at least one new patient, comprising: providing at least one representation of the at least one new patient's at least one organ at risk relative to at least one target; searching for at least one prior treatment plan for at least one prior patient with at least one similar representation; and reviewing the at least one prior treatment plan for the at least one prior patient in order to determine whether the at least one new treatment plan can be improved based on information in the at least one prior treatment plan.