Related Experiment Video
Updated: Aug 4, 2026
12:10
Single Particle Electron Microscopy Reconstruction of the Exosome Complex Using the Random Conical Tilt Method
Published on: March 28, 2011
Implementation and evaluation of an ordered subsets reconstruction algorithm for transmission PET studies using
V Bettinardi1, S Alenius, P Numminen
1Department of Nuclear Medicine, Scientific Institute H.San Raffaele, University of Milano-Bicocca, INB-CNR, Via Olgettina 60, 20132, Milan, Italy. bettinardi.valentino@hsr.it
European Journal of Nuclear Medicine and Molecular Imaging
|January 29, 2003
Summary
A new ordered subsets algorithm using median root prior (OS-MRP-TR) enables fast, low-noise transmission scans for positron emission tomography (PET) whole-body studies. This method allows for accurate attenuation correction with significantly reduced scan times.
Area of Science:
- Medical Imaging
- Nuclear Medicine
- Image Reconstruction
Background:
- Accurate attenuation correction is crucial for quantitative positron emission tomography (PET).
- Low count statistics in transmission (TR) scans can degrade image quality and attenuation correction accuracy.
- Traditional filtered back-projection (FBP) requires long scan times for adequate TR data.
Purpose of the Study:
- To implement and evaluate an ordered subsets (OS) reconstruction algorithm incorporating median root prior (MRP) and inter-update median filtering for low count statistics TR scans.
- To assess the performance of the OS-MRP-TR algorithm in terms of accuracy, noise reduction, and image quality compared to FBP.
- To validate the use of OS-MRP-TR reconstructed images for attenuation correction in whole-body (WB) PET studies.
Main Methods:
- Developed an ordered subsets reconstruction algorithm with median root prior (OS-MRP-TR).
- Evaluated the algorithm using experimental phantoms and patient data under various TR scan times (1 hour down to 1 minute).
- Assessed performance by comparing mean value of attenuation coefficient (MVAC) and coefficient of variation (CV) against FBP reconstructions.
Main Results:
- The OS-MRP-TR algorithm demonstrated stable solutions for MVAC.
- Achieved MVAC within 5% of FBP (1-hour scan) even with a 1-minute TR scan.
- Effectively reduced noise, showing lower CV for 1-minute OS-MRP-TR scans compared to 1-hour FBP scans.
- EM images corrected using 1-minute OS-MRP-TR scans showed mean counts within 3% of those corrected using 1-hour scans.
- Preserved good image quality for both TR and emission (EM) data.
Conclusions:
- The OS-MRP-TR algorithm is highly suitable for whole-body PET studies.
- Enables significant reduction in TR scan acquisition time (down to 1 minute) without compromising image quality.
- Facilitates accurate attenuation correction using low-noise TR images reconstructed from short scans.
