Multi-Modality Image Registration via Reference Object Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Multi-modality imaging systems often suffer from errors in reconstruction centers of functional and structural images, leading to reduced registration accuracy, which hampers the precise determination of lesion locations and diagnostic reliability.
Innovation Solution
A method and system that acquire and process images using a registration model with reference objects, determining reference points and mapping data to establish a mapping relation between coordinate positions in different images, and calculating registration parameters to correct for errors and enhance alignment accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multi-modality imaging systems generate functional and structural images separately, then more detailed diagnostic information is provided, but errors in reconstruction centers reduce registration accuracy
Solution Approach 1:
The patent introduces a registration model containing reference objects as an intermediary between functional and structural images. This registration model serves as a mediator to establish accurate spatial correspondence between the two image types, compensating for reconstruction center errors and enabling precise registration while maintaining the benefits of multi-modality imaging.
2Loss of information
If separate functional and structural images are used for diagnosis, then more comprehensive information is obtained, but lesion location determination precision is reduced due to registration errors
Solution Approach 1:
The registration model with reference objects acts as an intermediary framework that preserves all diagnostic information from both functional and structural images while enabling precise lesion location determination through accurate spatial mapping between the two modalities.
Solution Approach 2:
The patent segments the registration process into distinct components: reference objects are identified separately in both functional and structural images, their coordinates are independently determined, and then mapping relationships are established between corresponding reference points, enabling precise lesion localization through this segmented approach.
3Device complexity
If traditional registration methods are used without correction, then the process is simpler, but reconstruction center errors persist and reduce registration accuracy
Solution Approach 1:
The patent applies preliminary action by first identifying and determining coordinates of reference objects in both functional and structural images before performing the actual registration. This preliminary step of establishing reference point mappings corrects reconstruction center errors in advance, enabling accurate registration without significantly increasing overall process complexity.
Data Source
AI summary
The present disclosure relates to a method and system for registering multi-modality images. The method may include: acquiring a first image relating to one or more reference objects; acquiring a second image relating to the one or more reference objects; determining a set of reference points based on the one or more reference objects; determining a set of mapping data corresponding to the set of reference points in the first image and the second image; and determining one or more registration parameters by comparing a plurality of back-projection errors generated in a plurality of iterations that are performed based on the set of mapping data.