Live PET Image Reconstruction via Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional positron-emission-tomography (PET) imaging is time and resource-consuming, leading to delayed image availability, which limits its use in conjunction with interventional procedures and does not provide suitable indications of temporal tracer distribution.
Innovation Solution
The method involves direct localization of PET data to Cartesian spatial locations for near real-time image generation during PET scanning, allowing for live interventional procedures. Additional PET data from different body regions is combined with existing data to create a field of view of an entire moving scan, with regions being gradually filled in as the scan progresses.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional PET reconstruction methods are used, then complete three-dimensional image volumes are obtained, but image availability is delayed and reconstruction is time-consuming
Solution Approach 1:
The patent divides the complete PET scan into multiple segments or frames acquired at different time points. Each segment is reconstructed separately into a partial image volume, allowing progressive display of image data as each segment is processed, rather than waiting for the complete scan to finish
Solution Approach 2:
The system performs preliminary reconstruction of early acquired PET segments before the complete scan is finished. This allows image data to be prepared and made available in advance, enabling interventional procedures to begin while later segments are still being acquired and reconstructed
2Measurement precision
If conventional PET reconstruction methods are used, then complete image volumes are obtained, but the process is resource-consuming
Solution Approach 1:
The reconstruction process is segmented into multiple smaller tasks corresponding to different time frames or body regions. Each segment can be processed independently and in parallel, reducing the computational burden on any single processing unit and improving overall reconstruction efficiency
Solution Approach 2:
The system performs partial reconstruction of only the necessary portions of the PET data at each stage rather than completing the full reconstruction. This allows useful image data to be generated with fewer computational resources while maintaining sufficient diagnostic quality for interventional guidance
3Area of stationary object
If PET scanning continues to acquire complete data, then comprehensive image coverage is achieved, but real-time imaging capability is lost
Solution Approach 1:
The field of view is divided into multiple regions that are scanned and reconstructed sequentially or in parallel. Each region generates image data independently at high speed, allowing real-time display of covered regions while the scanner continues acquiring data in other regions, ultimately achieving complete coverage
Solution Approach 2:
The system preliminarily reconstructs and displays image data from already-acquired segments before the complete scan is finished. This creates a dynamic, progressively updating image that provides real-time feedback while the remaining scan data is being collected and processed
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enables near real-time image generation during PET scanning, allowing for live interventional procedures and providing timely and useful images based on PET data.
Implementation Method 1
the radioisotope tracer emits positrons which annihilate with electrons to produce gamma rays
Implementation Method 2
positrons which annihilate with electrons to produce gamma rays
Implementation Method 3
A detector system located outside the body detects the emitted gamma rays
Data Source
AI summary
A system and method include localization of a first frame of positron emission tomography data acquired by an imaging device to a first frame of Cartesian data, generation of a first Cartesian image volume based on the first frame of Cartesian data, display of the first Cartesian image volume, localization of a second frame of positron emission tomography data acquired by the imaging device to a second frame of Cartesian data, generation of a second Cartesian image volume based on the second frame of Cartesian data, and display of the combined Cartesian image volume.


