Fast formation of statistically reliable FDG parametric images based on clustering and principal components

• Published in: Physics in medicine & biology Vol. 47; no. 3; pp. 455 - 468
• Main Authors: Kimura, Y, Senda, M, Alpert, N M
• Format: Journal Article
• Language: English
• Published: Bristol IOP Publishing 07.02.2002; Institute of Physics
• Subjects: Algorithms; Biological and medical sciences; Brain - pathology; Cluster Analysis; Humans; Investigative techniques, diagnostic techniques (general aspects); Kinetics; Medical sciences; Models, Theoretical; Nervous system; Radionuclide investigations; Software; Time Factors; Tomography, Emission-Computed - instrumentation; Tomography, Emission-Computed - methods; Radionuclide study; Cluster analysis; Human; Nervous system diseases; Parameter estimation; Alzheimer disease; Optimization; Cerebral disorder; Parametric method; Image quality; Radiation dose; Central nervous system disease; Degenerative disease; Arithmetic mean; Pharmacokinetics; Medical application; Principal component analysis; Brain (vertebrata); Positron; Signal to noise ratio; Emission tomography

Formation of parametric images requires voxel-by-voxel estimation of rate constants, a process sensitive to noise and computationally demanding. A model-based clustering method for a two-parameter model (CAKS) was extended to the FDG three-parameter model. The concept was to average voxels with simi...