Advanced search
Start date
Betweenand

Algorithms for tomographic reconstruction: optimization, restoration, quantification and clinical application

Abstract

Tomographic reconstruction created a major impact in Medicine. It has allowed non invasive visualization of anatomic structures, metabolism and function of human body. Over the last 30 years, the progresses on tomographic reconstruction techniques have been enormous, both in terms of quality and performance, as well as on dynamic and tridimensional techniques [Udupa, 2000]. However, there are still several open challenges in this field, mainly in emission tomographic techniques, such as: (a) truly quantitative SPECT (Single Photon Emission Computed Tomography) and PET (Positron Emission Tomography); (b) truly 40 tomographic reconstruction algorithms for dynamic structures; (c) robust tridimensional segmentation and quantification; and (d) objective evaluation of tomographic methods. Emission tomography has the unique ability to obtain functional and metabolic information through the use of radiopharmaceuticals. This information may indicate changes in biological processes, yielding early detection of diseases, much before anatomical changes take place. In some situations, even without any physical symptom or anatomical indication, the emission images are able to show altered functions. The main objective of this project is to sum up efforts and knowledge of several groups that have been working with several aspects of tomography for a long time. The investigations basically include: a) practical and new optimized reconstruction algorithms; b) alternative image restoration algorithms; c) robust volume quantification and segmentation processes; and d) clinical applications. Specific objectives are: Research, development and implementation of efficient reconstruction algorithms (20 and 3D) for Nuclear Medicine, aiming the reduction of absorbed doses as well as the image acquisition and processing time; Research of methods for quantitative tomographic reconstructions in Nuclear Medicine dealing with correction/compensation and restoration of images; Research on truly 40 reconstruction algorithms for dynamic structures; Research, development and evaluation of the whole reconstruction process, including restoration and quantification, through physical and numerical simulation in 20, 3D and 40. Proposal of clinical protocols and applications based on quantitative SPECT and PET. The relevance of the present proposal is the coordinated and consistent investigation of solutions for important questions in Emission Tomography... (AU)

Scientific publications (6)
(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)
MIQUELES, EDUARDO X.; DE PIERRO, ALVARO RODOLFO. On the Inversion of the xfct Radon Transform. STUDIES IN APPLIED MATHEMATICS, v. 127, n. 4, p. 394-419, NOV 2011. Web of Science Citations: 3.
ZIBETTI, MARCELO V. W.; DE PIERRO, ALVARO R. A New Distortion Model for Strong Inhomogeneity Problems in Echo-Planar MRI. IEEE TRANSACTIONS ON MEDICAL IMAGING, v. 28, n. 11, p. 1736-1753, NOV 2009. Web of Science Citations: 4.
HELOU NETO, ELIAS SALOMAO; DE PIERRO, ALVARO RODOLFO. INCREMENTAL SUBGRADIENTS FOR CONSTRAINED CONVEX OPTIMIZATION: A UNIFIED FRAMEWORK AND NEW METHODS. SIAM JOURNAL ON OPTIMIZATION, v. 20, n. 3, p. 1547-1572, 2009. Web of Science Citations: 23.
DE PIERRO, ALVARO R.; CREPALDI, FABIANA. Activity and attenuation recovery from activity data only in emission computed tomography. COMPUTATIONAL & APPLIED MATHEMATICS, v. 25, n. 2-3, p. 205-227, 2006. Web of Science Citations: 14.
SANTOS, REGINALDO J.; DE PIERRO, ALVARO R. The effect of the nonlinearity on GCV applied to Conjugate Gradients in Computerized Tomography. COMPUTATIONAL & APPLIED MATHEMATICS, v. 25, n. 1, p. 111-128, 2006. Web of Science Citations: 3.
WEI‚ M.; DE PIERRO‚ A.R.; YIN‚ J. Error estimates for two filters based on polynomial interpolation for recovering a function from its Fourier coefficients. NUMERICAL ALGORITHMS, v. 35, n. 2, p. 205-231, 2004.

Please report errors in scientific publications list by writing to: cdi@fapesp.br.