Limited Field of View Localization for Magnetic Resonance Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current whole-body MR scanners face challenges in accurately positioning and scanning deep-seated targets, such as the prostate, due to limitations in initial mechanical alignment and the need for large homogeneous scanning volumes.
Innovation Solution
The method involves performing multiple scout scans of different regions of the patient to accurately locate the target, using a machine-learned model to determine the target's position from the scout scans, and configuring the MR scanner for diagnostic imaging based on this position.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If a large homogeneous scanning volume is used for whole-body MR scanning, then the field of view is increased and mechanical alignment is simplified, but the system complexity and scan time for deep-seated targets increase
Solution Approach 1:
The patent divides the scanning process into multiple scout scans of different regions instead of requiring a single large homogeneous field of view. This segmentation allows the system to localize deep-seated targets using multiple smaller scans, reducing the requirement for a large homogeneous scanning volume while maintaining the ability to locate targets throughout the body.
2Measurement precision
If multiple scout scans of different regions are performed, then the localization accuracy for deep-seated targets is improved, but the scan time increases
Solution Approach 1:
The patent performs preliminary scout scans of different regions to localize the target before the main diagnostic scan. This preliminary localization action allows the system to quickly identify target positions using machine-learned models, enabling efficient planning of the subsequent diagnostic scan and reducing overall scan time despite multiple scout scans.
Solution Approach 2:
The patent uses machine-learned models to analyze scout scan data and determine target positions, changing the parameter of localization method from traditional mechanical alignment to AI-based analysis. This allows faster and more accurate localization from multiple scout scans compared to conventional methods.
3Volume of stationary object
If a compact MR system with smaller scanning volume is used, then the system size is reduced, but the ability to position deep-seated targets within the scanning volume is compromised
Solution Approach 1:
The patent introduces machine-learned models as an intermediary to bridge the gap between the compact scanner's limited field of view and the need to locate deep-seated targets. The AI model analyzes scout scans from different regions to determine target positions, enabling effective target positioning even when the physical scanning volume is small.
Data Source
AI summary
Rather than the whole-body scout scan, multiple scout scans of different regions are used to locate the target for magnetic resonance scanning. A diagnostic scan may be planned in as little time as possible based on localization using multiple scout scans of different regions of the patient. The planning of the different regions may be optimized to minimize the number and/or time for localizing the target.