Multi-Contrast MRI Reconstruction Using Joint and Individual Information Terms
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current multi-contrast MRI reconstruction techniques face challenges in reducing scan times and reconstruction time, as well as feature leakage across contrast images, which hinders clinical application and compatibility with parallel imaging techniques.
Innovation Solution
The proposed method utilizes both individual and joint information terms in compressive sensing reconstruction, allowing for simultaneous reconstruction of images with unequal acceleration rates and handling of sensitivity maps, thereby maximizing common information while preserving individual features and preventing feature leakage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If joint reconstruction techniques are used to process multi-contrast images, then common information across different contrast images can be utilized to improve reconstruction quality, but reconstruction time increases significantly
Solution Approach 1:
The patent segments the reconstruction process into separate processing stages for different contrast images while maintaining joint optimization. Each contrast image is processed individually through the optimization pipeline, but the segmentation allows parallel computation and independent tuning of parameters for each contrast type, thereby reducing overall reconstruction time while preserving the benefits of joint information utilization.
Solution Approach 2:
The patent applies partial joint processing by selectively combining information from multiple contrast images only where beneficial, rather than fully joint processing all images simultaneously. This partial action approach utilizes common information to improve reconstruction quality in regions where contrasts share structural information, while avoiding the computational burden of complete joint reconstruction across all images.
2Productivity
If joint reconstruction is employed to reduce scan times, then fewer data samples are required, but features may leak across different contrast images
Solution Approach 1:
The patent applies local quality by allowing different reconstruction parameters, regularization strengths, and processing strategies for different contrast images and different regions within images. Each contrast image receives customized processing that preserves its unique feature characteristics while still benefiting from joint information. This localized approach prevents feature leakage by maintaining distinct processing identities for each contrast type.
Solution Approach 2:
The patent introduces intermediary processing steps and separate optimization variables for each contrast image that act as mediators between the joint reconstruction framework and individual contrast features. These intermediaries preserve the uniqueness of each contrast image's features while enabling information exchange across contrasts, thereby preventing direct feature leakage while maintaining the advantages of joint processing.
3Loss of time
If conventional imaging techniques are used, then reconstruction time is negligible, but scan times are longer due to multiple protocols
Solution Approach 1:
The patent merges multiple contrast image reconstructions into a unified optimization framework that processes all contrasts simultaneously. This combining approach allows the system to leverage correlations and shared information across different contrast images, enabling significant reduction in scan time by acquiring fewer samples per contrast while maintaining reconstruction quality comparable to or better than conventional sequential imaging.
Data Source
AI summary
A method to reconstruct images from raw MRI data, and recover an image by solving an optimization problem. Hence, the method is an image reconstruction system for image recovery contrary to similar methods, the method reconstructs all images using both joint and individual objective functions. The method includes: acquiring with the MRI system, multiple images under an influence of different contrast mechanisms, wherein the different contrast mechanisms belong to a same anatomy, solving an optimization problem with an optimization algorithm using both joint and individual objective functions simultaneously for each of multiple image contrasts or a subset of the multiple image contrasts, to reconstruct the multiple images from acquired data.