Formation of Parametric Images in Positron Emission Tomography Using a Clustering-Based Kinetic Analysis With Statistical Clustering

Abstract

A method is proposed for forming parametric images in position emission tomography, using clustering kinetic analysis. To overcome the dual problems experienced in voxel-based data, of signal noise and the very long computational time, the data are clustered before parameter estimation, and then estimation procedure is applied to the averaged data in each cluster. Using this algorithm, PET data are optimally clustered, depending on the noise that is present, by hierarchically applying a statistical-clustering algorithm based on Mixed Gaussian model. In a computer simulation, the proposed method correctly clustered noise-contaminated data. Applying the proposed algorithm to (18)F-FDG clinical data, physiologically acceptable parametric images of glucose metabolism in a brain were obtained in a practical calculation time.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Oct 25, 2001
Accession Number
ADA410339

Entities

People

  • Keiichi Oda
  • Kenji Ishii
  • Yasuhiro Noshi
  • Yuichi Kimura

Tags

DTIC Thesaurus Topics

  • Abstracts
  • Algorithms
  • Biomedical Engineering
  • Clustering
  • Computer Simulations
  • Computers
  • Data Sets
  • Emission
  • Engineering
  • Noise
  • Positron Emission Tomography
  • Positron Emissions
  • Positrons
  • Probability
  • Simulations
  • Statistical Algorithms
  • Tomography

Fields of Study

  • Physics

Readers

  • Medical Imaging.
  • Neural Network Machine Learning.
  • Regression Analysis.