Digital Image Processing

Open Access Open Access  Restricted Access Subscription Access
Open Access Open Access Open Access  Restricted Access Restricted Access Subscription Access

Analysis of Qualitative Algorithms in Iterative Reconstruction of PET Data


Affiliations
1 VLB Janakiammal College of Engineering and Technology, Coimbatore, India
2 EEE Department, VLB Janakiammal College of Engineering and Technology, Coimbatore, India
     

   Subscribe/Renew Journal


This paper makes a comparison between the slow globally convergent list-mode COSEM and the fast but non-convergent classical list-mode OSEM methods and produce a way to get a fast and effective COSEM called the enhanced COSEM (ECOSEM). The former globally convergent algorithm is based on the ordered sets expectation maximization algorithm for binned data but has several extensions that make it suitable for two planar detector tomography that are rotating but experience certain limitations. The image estimate measured from the PET scanner is incrementally updated for each coincidence event measured. The events obtained are used as soon as possible, that improves the current image estimate, and therefore the convergence speed towards the maximum-likelihood is accelerated. We provide a simulated result to overcome those limitations and to reconstruct the image with maximum-likelihood.

Keywords

Maximum-Likelihood (Ml), Positron Emission Tomography (PET), Row-Action Maximum-Likelihood Algorithm (RAMLA), Expectation-Maximization (EM), Event-By-Event (EBE), Ordered Subsets (OS), Bayesian, Filtered Back Projection (FBP), Maximum A-Posteriori (MAP), Penalized Weighted Least-Squares (PWLS).
User
Subscription Login to verify subscription
Notifications
Font Size

Abstract Views: 241

PDF Views: 2




  • Analysis of Qualitative Algorithms in Iterative Reconstruction of PET Data

Abstract Views: 241  |  PDF Views: 2

Authors

V. Dhineshkumar
VLB Janakiammal College of Engineering and Technology, Coimbatore, India
S. Gobhinath
VLB Janakiammal College of Engineering and Technology, Coimbatore, India
R. Hariprashad
VLB Janakiammal College of Engineering and Technology, Coimbatore, India
K. Prabhakaran
VLB Janakiammal College of Engineering and Technology, Coimbatore, India
S. Vimalraj
EEE Department, VLB Janakiammal College of Engineering and Technology, Coimbatore, India

Abstract


This paper makes a comparison between the slow globally convergent list-mode COSEM and the fast but non-convergent classical list-mode OSEM methods and produce a way to get a fast and effective COSEM called the enhanced COSEM (ECOSEM). The former globally convergent algorithm is based on the ordered sets expectation maximization algorithm for binned data but has several extensions that make it suitable for two planar detector tomography that are rotating but experience certain limitations. The image estimate measured from the PET scanner is incrementally updated for each coincidence event measured. The events obtained are used as soon as possible, that improves the current image estimate, and therefore the convergence speed towards the maximum-likelihood is accelerated. We provide a simulated result to overcome those limitations and to reconstruct the image with maximum-likelihood.

Keywords


Maximum-Likelihood (Ml), Positron Emission Tomography (PET), Row-Action Maximum-Likelihood Algorithm (RAMLA), Expectation-Maximization (EM), Event-By-Event (EBE), Ordered Subsets (OS), Bayesian, Filtered Back Projection (FBP), Maximum A-Posteriori (MAP), Penalized Weighted Least-Squares (PWLS).