Motion Compensated MRI System with Dual-Bin Data Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Dynamic contrast enhanced magnetic resonance imaging faces challenges in capturing clear images due to subject motion, particularly during breathing, which corrupts data and complicates the imaging process as different processes occur on varying time scales, making it difficult to obtain high-quality images with contrast agent dynamics.
Innovation Solution
A medical imaging system and method that continuously acquires magnetic resonance data with motion signal values, bins data into predetermined motion and time bins, reconstructs reference images, and iteratively generates enhanced images using displacement vector fields to correct for motion and contrast changes, ensuring high-quality imaging despite subject movement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If continuous acquisition of magnetic resonance data is performed during subject motion, then temporal resolution is improved, but image quality deteriorates due to motion corruption
Solution Approach 1:
The patent segments the continuous magnetic resonance data into discrete data portions, each associated with specific motion signal values. By dividing the data acquisition into discrete segments that can be individually processed and corrected based on their motion characteristics, the system maintains high temporal resolution while enabling quality control through selective processing of motion-corrupted data
Solution Approach 2:
The patent introduces motion signal values as an intermediary parameter that mediates between the acquired magnetic resonance data and the final image reconstruction. These motion signals serve as a bridge to detect and quantify subject motion, enabling the system to compensate for motion effects and maintain image quality while continuing data acquisition during motion
2Loss of information
If discrete data portions are acquired with rotating k-space trajectory, then data coverage is improved, but motion artifacts increase due to different motion states across acquisitions
Solution Approach 1:
The patent implements feedback by continuously monitoring motion signal values associated with each discrete data portion and using this information to guide the reconstruction process. The motion signals provide feedback about subject state changes, allowing the system to adjust processing parameters and apply appropriate corrections to minimize motion artifacts while maintaining comprehensive data coverage
Solution Approach 2:
The patent applies preliminary action by associating motion signal values with each discrete data portion during acquisition. This preliminary tagging of motion information allows for subsequent retrospective correction and sorting of data based on motion states, enabling the system to recover usable information from data acquired during motion while maintaining comprehensive k-space coverage
3Manufacturing precision
If data is binned into predetermined motion bins, then motion compensation is improved, but temporal resolution deteriorates due to sorting by motion state
Solution Approach 1:
The patent adds another dimension to the data organization by creating a dual-binning structure: first binning by motion state to enable motion compensation, then binning by time to preserve temporal resolution. This two-dimensional sorting approach allows the system to simultaneously achieve motion correction and maintain temporal information, resolving the contradiction between motion compensation quality and temporal resolution
Data Source
AI summary
The invention provides for a medical imaging system (100, 300, 500) comprising a processor (104). Machine executable instructions cause the processor to: receive (200) magnetic resonance data (120) comprising discrete data portions (612) that are rotated in k-space; bin (202) the discrete data portions into predetermined motion bins (122) using a motion signal value; reconstruct (204) a reference image (124) for each of the predetermined motion bins; construct (206) a motion transform (126) between the reference images; bin (208) a chosen group (610) of the discrete data portions into a chosen time bin (128). Generate an enhanced image (130) for the chosen time bin using the chosen group fo the discrete data portions and the motion transform of each of the chosen group to correct the discrete data portions.