Nuclear Image Reconstruction via Breathing Model Motion Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current nuclear imaging techniques, such as SPECT and PET, face challenges in obtaining accurate images due to patient respiratory motion, which introduces motion artifacts and makes it difficult to provide correct morphological information for attenuation correction, especially since breath-holding is not feasible during long acquisition times.
Innovation Solution
A system that uses a breathing model to estimate patient motion and organ deformation, generating a time-varying attenuation map for real-time correction of nuclear images, allowing for efficient acquisition of sparse motion information at key locations and adapting it to provide high-quality nuclear images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If traditional nuclear imaging is performed without motion correction, then acquisition time is reduced, but image quality deteriorates due to motion artifacts
Solution Approach 1:
The system performs preliminary motion estimation using a breathing model before nuclear image acquisition, and prepares correction transformations in advance. This allows motion compensation to be applied during reconstruction without extending the actual nuclear imaging acquisition time, thus maintaining image quality while avoiding time loss.
Solution Approach 2:
The system creates a computational model (breathing model) that replicates patient respiratory patterns. This model copy is then used to generate correction transformations that compensate for actual patient motion during acquisition, avoiding the need for lengthy breath-holding protocols while maintaining image quality.
2Measurement precision
If detailed motion information is acquired across the entire region of interest, then motion correction accuracy is improved, but device complexity and acquisition time increase
Solution Approach 1:
The system divides the complex task of motion tracking into segments: (1) sparse motion information acquisition at limited locations using simple sensors, (2) breathing model-based estimation for the entire region, and (3) correction transformation generation. This segmentation reduces device complexity while maintaining correction accuracy through the use of a computational breathing model.
Solution Approach 2:
The breathing model acts as an intermediary between sparse motion measurements and full-field correction transformations. Instead of directly measuring motion everywhere, the system uses the breathing model to translate limited sensor data into comprehensive correction information, reducing the need for complex direct measurement systems.
3Measurement precision
If breath-holding is used during acquisition, then motion artifacts are reduced, but patient comfort and feasibility deteriorate due to long acquisition time
Solution Approach 1:
The system transitions from static breath-holding protocols to dynamic real-time motion compensation. The breathing model continuously adapts to the patient's natural respiratory pattern during acquisition, generating time-varying correction transformations that compensate for motion without requiring the patient to alter their breathing, thus improving comfort while maintaining image quality.
Solution Approach 2:
The system implements feedback by continuously monitoring sparse motion signals and using the breathing model to generate real-time correction transformations. This closed-loop approach allows the system to respond to actual patient motion during acquisition, maintaining image quality without requiring breath-holding, thereby improving patient comfort and feasibility.
Data Source
Figure 1
Figure 2
AI summary
A system is provided for obtaining a nuclear image of a moving object. The system comprises an input (14), a processing unit (15) and an output (17). The input (14) is provided for receiving a nuclear image and morphological images of the object. The processing unit (15) is configured to process the morphological images to obtain sparse motion information of the object, to use the sparse motion information and a motion model for obtaining estimated motion information about the object, and to generate a motion- corrected nuclear image based on the estimated motion information and the acquired nuclear image. The output (17) provides the corrected nuclear image.