Automated Anatomy Delineation via Landmark Mapping and Spline Interpolation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current medical imaging systems, particularly in radiation treatment planning, face challenges in accurately delineating anatomical structures due to limited soft tissue contrast in CT images and variability in MRI data, making reliable organ boundary discrimination difficult and non-reproducible.
Innovation Solution
A system and method that utilize a processor to detect anatomical landmarks in initial images, compare and map them to reference landmarks, adjust reference contours, and apply these adjustments to high-resolution images for improved therapy planning, employing techniques like spline interpolation for accurate contour adjustment and propagation across imaging modalities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If CT imaging is used for therapy planning, then the imaging process is fast and widely available, but soft tissue contrast is poor making organ boundary discrimination difficult
Solution Approach 1:
The patent combines CT and MRI imaging modalities to achieve both fast imaging (from CT) and superior soft tissue contrast (from MRI). The system integrates anatomical information from CT with functional and contrast information from MRI to create comprehensive therapy plans that overcome the limitations of either modality alone.
Solution Approach 2:
The patent uses automated segmentation algorithms and image registration techniques as intermediary processes to bridge CT and MRI data. These computational methods process and fuse the imaging data from both modalities, enabling accurate organ boundary discrimination while maintaining imaging speed.
2Measurement precision
If MRI is used for therapy planning, then soft tissue contrast is superior, but automated segmentation is difficult due to non-reproducible gray value distribution
Solution Approach 1:
The patent implements feedback mechanisms where the segmentation algorithm iteratively refines its results by comparing automated detections with manual expert annotations. This feedback loop allows the system to learn from discrepancies and improve segmentation accuracy while handling the complexity of MRI data with varying gray value distributions.
Solution Approach 2:
The patent employs multiple imaging parameters and contrast mechanisms in MRI (such as different pulse sequences and weighting) to enhance tissue differentiation. By optimizing and adjusting these parameters, the system improves soft tissue contrast while providing more consistent data for automated segmentation algorithms.
3Measurement precision
If manual anatomy delineation is performed, then accuracy can be high, but time consumption and subjectivity increase
Solution Approach 1:
The patent performs preliminary automated segmentation and contour generation before final review. This preliminary action creates a draft delineation that captures the majority of anatomical structures, reducing the time required for manual refinement while maintaining high accuracy through subsequent expert verification.
Solution Approach 2:
The patent enables the system to perform self-correction and self-refinement of segmentation results through iterative algorithms that automatically adjust contours based on image features and previously learned patterns. This self-service capability reduces reliance on manual intervention while maintaining delineation accuracy.
Data Source
AI summary
When delineating anatomical structures in a medical image of a patient for radiotherapy planning, a processor (18) detects landmarks (24) in a low-resolution image (e.g., MRI or low-dose CT) and maps the detected landmarks to reference landmarks (28) in a reference contour of the anatomical structure. The mapped landmarks facilitate adjusting the reference contour to fit the anatomical structure. The adjusted reference contour data is transformed and applied to a second image using a thin-plate spline, and the adjusted high-resolution image is used for radiotherapy planning.