Tracker-Less Image Registration Using Semantic Landmark Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image fusion methods between dynamic and static imaging modalities require additional tracking hardware, which increases costs and processing time, making it inefficient for high frame rate applications.
Innovation Solution
A tracker-less method for registering and combining images from different datasets using semantic and geometric descriptors to match landmarks, reducing the need for additional hardware and speeding up the image registration process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional image fusion methods using tracking hardware are used, then image registration accuracy is improved, but device complexity and cost increase
Solution Approach 1:
The patent extracts and removes the tracking hardware component from the image fusion system. Instead of using external trackers to monitor probe position, the invention uses purely image-based landmark detection and matching algorithms to achieve registration, thereby eliminating the complexity and cost of additional tracking devices while maintaining registration accuracy
Solution Approach 2:
The image registration system performs self-service by using its own image data to determine probe position and orientation. The system detects anatomical landmarks within the ultrasound images themselves and uses these landmarks to automatically register the dynamic ultrasound images with static reference images, without requiring external tracking systems
2Measurement precision
If traditional tracking-based registration methods are used, then registration accuracy is improved, but processing time increases
Solution Approach 1:
The patent applies preliminary action by pre-identifying and storing landmark positions and characteristics in the static reference images before the actual image fusion process. This pre-processing allows the dynamic ultrasound images to be rapidly registered by simply matching landmarks against the pre-established reference framework, significantly reducing real-time processing requirements
Solution Approach 2:
The registration process is segmented into distinct stages: landmark detection, landmark matching, and transformation calculation. This segmentation allows each stage to be optimized independently, with landmark detection using efficient image processing algorithms and landmark matching using geometric constraints, thereby reducing overall processing time while maintaining accuracy
3Productivity
If high frame rate dynamic imaging is used, then real-time monitoring capability is improved, but image quality and resolution decrease
Solution Approach 1:
The patent merges the advantages of both high frame rate dynamic imaging and high resolution static imaging by fusing ultrasound images with reference images from other modalities. The dynamic ultrasound provides real-time temporal information at high frame rates, while the static reference images contribute high spatial resolution and anatomical detail, creating a composite image that benefits from both characteristics
Solution Approach 2:
The landmark-based registration system acts as an intermediary that bridges the gap between dynamic and static images. By establishing accurate geometric relationships through landmark matching, the system allows high frame rate ultrasound images to be precisely overlaid on high resolution reference images, enabling real-time monitoring without sacrificing image quality
Data Source
AI summary
A method for tracker-less registering images of the same object generated from two different image datasets, and particularly from two different image datasets acquired by means of two different image acquisition methods and/or techniques comprising:identifying landmarks in the image or images of the two dataset;associating to each landmark a univocal semantic description comprising semantic labels describing at least one feature of the landmark and/or geometrical labels describing at least one or some of the geometric relationships between the landmarks identified and/or geometrical labels relating to the shape and/or orientation of the said identified landmarks;considering two images of the said two data sets as being images along image slicers or image planes registered one with the other when the semantic descriptions related to the landmarks present in the said images are matching one with the other.