Denoising of high resolution small animal 3D PET data using the non-subsampled Haar wavelet transform

PET allows functional imaging of the living tissue. However, one of the most serious technical problems affecting the reconstructed data is the noise, particularly in images of small animals. In this paper, a method for high-resolution small animal 3D PET data is proposed with the aim to reduce the...

Full description

Saved in:

Bibliographic Details
Published in:	Nuclear instruments & methods in physics research. Section A, Accelerators, spectrometers, detectors and associated equipment Vol. 784; pp. 581 - 584
Main Authors:	Ochoa Domínguez, Humberto de Jesús, Máynez, Leticia O., Vergara Villegas, Osslan O., Mederos, Boris, Mejía, José M., Cruz Sánchez, Vianey G.
Format:	Journal Article
Language:	English
Published:	Elsevier B.V 01-06-2015
Subjects:	Haar wavelet transform Image denoising PET image denoising Haar wavelet transform PET image denoising Image denoising
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	PET allows functional imaging of the living tissue. However, one of the most serious technical problems affecting the reconstructed data is the noise, particularly in images of small animals. In this paper, a method for high-resolution small animal 3D PET data is proposed with the aim to reduce the noise and preserve details. The method is based on the estimation of the non-subsampled Haar wavelet coefficients by using a linear estimator. The procedure is applied to the volumetric images, reconstructed without correction factors (plane reconstruction). Results show that the method preserves the structures and drastically reduces the noise that contaminates the image.
ISSN:	0168-9002 1872-9576
DOI:	10.1016/j.nima.2014.12.109