Virtual Image Atlas for Patient Tracking in Radiation Therapy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current radiation treatment methods face challenges in accurately tracking patient position and orientation, particularly for soft tissue targets, due to limitations in external marker systems and invasive fiducial marker procedures, which can fail to account for internal displacements and soft tissue movements during treatment.
Innovation Solution
The use of a virtual image 'atlas' for deformable registration between pre-operative and intra-operative images, allowing for accurate tracking of patient position and orientation without additional radiation exposure, by registering CT and ultrasound images using a combined atlas that includes both modalities, enabling precise alignment of radiation beams.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If external markers are used for tracking, then tracking can be performed without invasive procedures, but the markers cannot adequately reflect internal displacements caused by breathing motion
Solution Approach 1:
The patent introduces an intermediary computational model (deformable registration algorithm) that maps the relationship between external skin markers and internal target tissue. This intermediary model compensates for the discrepancy between external and internal motion, allowing external markers to indirectly track internal displacements with high precision without requiring invasive implantation.
Solution Approach 2:
The patent creates a virtual copy of the internal target tissue's motion characteristics through deformable registration algorithms. This computational copy replicates the internal tissue's displacement patterns based on external marker data, enabling accurate tracking of internal motion without direct physical contact with the target tissue.
2Measurement precision
If implanted fiducial markers are used, then accurate tracking of internal target position can be achieved, but invasive implantation procedures are required
Solution Approach 1:
The patent extracts the essential tracking function from the physical fiducial marker and relocates it to a computational algorithm. By removing the need for physical implantation and replacing it with a deformable registration model that processes external marker data, the system achieves accurate target tracking without the harmful invasive procedures associated with traditional fiducial markers.
Solution Approach 2:
The patent replaces the mechanical system of implanted fiducial markers with a computational system based on deformable registration algorithms. This substitution eliminates the need for surgical implantation procedures while maintaining or improving tracking accuracy through mathematical modeling of tissue deformation and marker-to-target relationships.
3Measurement precision
If bone-implanted fiducial markers are used, then high tracking accuracy can be achieved, but difficult and painful invasive procedures are required
Solution Approach 1:
The patent removes the invasive bone implantation step entirely by extracting the tracking function to a computational level. The deformable registration algorithm processes data from non-invasive external markers to achieve tracking accuracy previously only attainable through bone-implanted fiducials, thereby eliminating the difficult and painful surgical procedures.
4Ease of operation
If fiducial markers are implanted in soft tissue, then tracking can be performed without bone procedures, but the markers may not provide accurate results for soft tissue target movement
Solution Approach 1:
The patent segments the tracking problem into two distinct components: external marker placement (simple, non-invasive) and internal tissue deformation modeling (complex, computational). By separating the physical measurement from the computational interpretation, the system achieves both ease of operation and high precision for soft tissue targets through the deformable registration algorithm that accounts for tissue-specific deformation patterns.
Data Source
AI summary
An apparatus and method of tracking a patient using a virtual image.