MRI Coil Sensitivity Map Updates for Motion-Aware Reconstruction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The acquisition of k-space data for magnetic resonance imaging is time-consuming, and existing parallel imaging techniques like SENSE struggle with joint optimization reconstructions when motion parameters are determined simultaneously, leading to impaired image quality due to missing coil sensitivity information during subject movement.
Innovation Solution
A medical system and method that includes a motion detection component to determine subject position, acquiring initial and subsequent coil sensitivity maps as needed, and using joint optimization to reconstruct magnetic resonance images with a collection of available coil sensitivity maps, adjusting for subject motion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If parallel imaging techniques like SENSE are used to accelerate MRI acquisition, then imaging speed is improved, but image quality deteriorates when subject motion occurs during scanning
Solution Approach 1:
The system dynamically updates coil sensitivity maps during the scanning process based on detected subject motion. Instead of using a static sensitivity map acquired before scanning, the system acquires updated sensitivity maps at intermediate time points and selects or combines them based on the subject's position, making the sensitivity map adaptive to motion conditions
Solution Approach 2:
The system implements a feedback loop where subject position is continuously monitored during scanning, and this position information is used to determine which coil sensitivity map to apply or how to combine multiple sensitivity maps. The motion detection feedback enables real-time adjustment of the reconstruction parameters to maintain image quality despite subject movement
2Measurement precision
If coil sensitivity maps are acquired before scanning to enable parallel imaging, then reconstruction accuracy is improved, but the system cannot adapt to subject motion during scanning
Solution Approach 1:
The system acquires multiple coil sensitivity maps at different time points before and during the scanning process in advance. These pre-acquired sensitivity maps are stored and later selected or combined based on the actual subject position during reconstruction, allowing the system to have adaptive capability without requiring real-time updates during critical acquisition phases
Solution Approach 2:
The system changes the temporal parameter of coil sensitivity map acquisition by obtaining sensitivity maps at multiple different time points rather than a single pre-scan time point. This creates a set of sensitivity maps corresponding to different subject positions, enabling the reconstruction algorithm to select the most appropriate sensitivity map based on actual motion conditions
3Manufacturing precision
If multiple coil sensitivity maps are acquired to account for subject motion, then motion artifact reduction is improved, but scanning time increases
Solution Approach 1:
The system acquires more coil sensitivity maps than the minimum single pre-scan map, obtaining additional maps at intermediate time points. However, not all acquired sensitivity maps are necessarily used in every reconstruction - the system selectively applies or combines them based on subject motion, achieving better motion compensation without proportionally increasing total scan time
Solution Approach 2:
The system acquires coil sensitivity maps at periodic intervals during the scanning process rather than continuously. This periodic acquisition strategy balances the need for motion compensation data with the constraint of total scanning time, obtaining sensitivity maps at strategically spaced time points that capture subject motion without excessive time penalty
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Disclosed herein is a method of operating a medical system (100), comprising: determining (200) initial position data (146) using a motion detection component, wherein the initial position data (146, 152) is descriptive of a subject within an imaging zone; acquiring (202) an initial coil sensitivity map (148) by controlling a magnetic resonance imaging system with coil sensitivity map pulse sequence commands; acquiring (204) imaging k-space data (150) according to a parallel imaging magnetic resonance imaging protocol by controlling the magnetic resonance imaging system with imaging pulse sequence commands; determining (206) subsequent position data (152) using the motion detection component; acquiring (208) at least one subsequent coil sensitivity map (154) by controlling the magnetic resonance imaging system with the coil sensitive map pulse sequence commands if the subsequent position data varies by more than a predetermined change since acquisition of a most recent coil sensitivity map; reconstructing (212) a magnetic resonance image (160, 402) based on the imaging k-space data and the at least one subsequent coil sensitivity map. A medical system is further presented, comprising a computational system configured to perform the method.