Related Experiment Video
Updated: Dec 15, 2025
07:52
Automated Rat Single-Pellet Reaching with 3-Dimensional Reconstruction of Paw and Digit Trajectories
Published on: July 10, 2019
14.9K
Ultra-Fast List-Mode Reconstruction of Short PET Frames and Example Applications
Matthew G Spangler-Bickell1, Timothy W Deller2, Valentino Bettinardi3
1Department of Radiology, University of Wisconsin, Madison, Wisconsin matthew.bickell@gmail.com.
Summary
This study introduces an infrastructure for ultra-fast positron emission tomography (PET) reconstructions, enabling rapid creation of short image frames. This facilitates applications like motion estimation and real-time visualization during scans.
Area of Science:
- Medical Imaging
- Nuclear Medicine
- Computational Science
Background:
- Standard clinical positron emission tomography (PET) reconstructions are time-consuming, hindering time-sensitive applications.
- Applications like data-driven motion estimation require numerous short frames, which are infeasible with current reconstruction speeds.
Purpose of the Study:
- To develop and present an infrastructure for ultra-fast list-mode PET reconstructions.
- To enable the creation of very short frames (≤1 second) for dynamic imaging applications.
Main Methods:
- Implemented and parallelized a fast ray-tracing time-of-flight projector.
- Optimized reconstruction parameters for speed, including limited iterations and no point-spread-function modeling or scatter correction.
- Utilized image registration for data-driven motion estimation from reconstructed frames.
Main Results:
- The infrastructure successfully generated images suitable for visualization, gating, and motion estimation.
- Demonstrated feasibility through case studies including brain, abdominal, and whole-body PET scans.
- Showcased data-driven motion estimation and correction, and visualization of physiological motions like respiration and cardiac activity.
Conclusions:
- The developed infrastructure enables rapid generation of short PET frames.
- This capability supports diverse applications, including real-time motion detection, improved gating, and enhanced visualization.
