Motion Compensated Imaging Using Patient-Adapted Anatomical Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current motion compensation techniques in imaging systems, such as SPECT/CT, face challenges in accurately capturing patient-specific motion patterns, particularly for non-rigid internal organ movements, and often require additional radiation or costly external devices, leading to suboptimal image quality and quantitative accuracy.
Innovation Solution
A method and system that adapt a generic anatomical model to individual patient anatomy from structural scans to generate a patient-specific motion model, which is then used for motion-compensated reconstruction of functional projection data, reducing the need for additional radiation and external devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If dynamic CT with gated reconstruction is used to derive motion models, then motion modeling accuracy is improved, but patient radiation dose increases and device complexity increases
Solution Approach 1:
The system performs preliminary adaptation of the generic anatomical model to the patient's specific anatomy using the structural scan data before deriving the motion model. This preliminary geometric adaptation ensures that the motion model is based on the correct anatomical structure, allowing accurate motion modeling to be derived from the functional scan without requiring additional dynamic CT imaging.
Solution Approach 2:
The system creates a patient-specific motion model by adapting a generic anatomical model to match the patient's unique anatomy and motion characteristics. This copying approach allows the system to capture patient-specific motion patterns without requiring additional radiation-exposing CT scans, using instead the adapted generic model combined with functional projection data.
2Measurement precision
If external devices like video cameras are used to track respiratory motion, then motion tracking capability is improved, but system cost and space requirements increase
Solution Approach 1:
The system uses the existing structural scan data and generic anatomical model to serve multiple functions: anatomical representation, motion modeling, and motion compensation. This multi-functionality eliminates the need for separate external motion tracking devices, as the same imaging system components are utilized for both structural imaging and motion characterization.
Solution Approach 2:
The imaging system performs motion tracking and motion model derivation using its own existing components (structural scan data and generic anatomical model) without requiring external devices. The system serves itself by utilizing the structural information already acquired during the imaging examination to characterize and compensate for motion.
3Ease of manufacture
If simple statistical models tracking center of mass are used, then ease of computation is improved, but motion modeling accuracy deteriorates
Solution Approach 1:
The system adapts the generic anatomical model to the patient's specific anatomy, creating a patient-specific model that captures local variations in motion characteristics. This local quality approach allows the motion model to accurately represent patient-specific anatomical variations and motion patterns while maintaining computational efficiency through the use of the adapted generic model framework.
Solution Approach 2:
The system changes the parameters of the generic anatomical model by adapting it to match the patient's specific anatomy using structural scan data. This parameter adaptation allows the motion model to capture patient-specific characteristics while maintaining the overall computational efficiency of the generic model framework, avoiding the need for completely new complex models for each patient.
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
A method includes performing a motion compensated reconstruction of functional projection data using a patient-adapted motion model, which is generated based on a generic anatomical motion model and imaging data from a structural scan. A system includes a first adapter (202) configured to adapt a generic anatomical model to structural image data, producing an adapted model, a forward projector (204) configured to forward project the adapted model, producing forward projected data, and a second adapter (206) configured to adapt the forward projected data to individual projections of projected data, which is used to generate the structural image data, producing a patient-adapted motion model.