Physiology-Compensated Image Reconstruction for Motion Artifacts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional medical imaging reconstruction methods assume a motionless subject, leading to reconstruction artifacts due to patient physiology changes during finite temporal resolution acquisitions, which degrade reconstruction accuracy and introduce uncertainty in downstream applications.
Innovation Solution
Incorporating prior knowledge of cardiac anatomy and motion into the reconstruction process to isolate and compensate for the influences of physiology, reformulating the reconstruction problem to address reconstruction artifacts by estimating independent influences of anatomy and physiology.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional reconstruction methods assume a motionless subject, then the reconstruction process is simple and fast, but reconstruction accuracy deteriorates due to artifacts from patient physiology changes
Solution Approach 1:
The patent segments the measurement signal into multiple representations, each corresponding to a different aspect (anatomy, motion, physiological process). This allows independent processing and compensation of each component, resolving the contradiction by enabling accurate physiology-compensated reconstruction through systematic decomposition rather than monolithic complex processing
Solution Approach 2:
The patent creates representations of the measurement signal before final reconstruction, allowing prior knowledge of cardiac anatomy and motion to be incorporated in advance. This preliminary decomposition enables artifact compensation during reconstruction without requiring complex real-time adjustments, improving accuracy while managing computational complexity
2Reliability
If physiology compensation is incorporated into reconstruction, then reconstruction accuracy improves, but computational time and processing complexity increase
Solution Approach 1:
By dividing the reconstruction problem into separate representations for anatomy, motion, and physiological processes, the patent enables parallel processing and selective computation. This segmentation allows the system to process only the necessary components based on clinical needs, reducing unnecessary computational time while maintaining accuracy
Solution Approach 2:
The patent reformulates the reconstruction problem by changing the parameter space from raw measurement signals to transformed representations that explicitly separate physiological influences. This parameter transformation enables more efficient algorithms to operate on simplified data structures, reducing computational complexity while improving reliability
3Object-affected harmful factors
If multiple representations of the measurement signal are created and processed, then artifact compensation improves, but device complexity and processing steps increase
Solution Approach 1:
The patent segments harmful physiological influences from the measurement signal by creating separate representations for each aspect. This allows targeted compensation of specific artifacts (e.g., cardiac motion, respiratory motion) independently, reducing overall complexity compared to handling all artifacts simultaneously through a single complex processing step
Data Source
AI summary
A computer-implemented method for medical measurement reconstruction may comprise: receiving a measurement acquisition signal; based on the received measurement acquisition signal, creating a plurality of representations of the measurement acquisition signal, wherein each of the plurality of representations relates to a different aspect of the measurement acquisition signal; modifying one or more of the plurality of representations; and generating an output signal including the modified one or more of the plurality of representations.