Exact parameter determination for Parkinson's disease diagnosis with PET using an algebraic approach

Hiroshi Yoshida, Koji Nakagawa, Hirokazu Anai, Katsuhisa Horimoto

Research output: Chapter in Book/Report/Conference proceedingConference contribution

5 Citations (Scopus)

Abstract

The mechanism of Parkinson's disease can be investigated at the molecular level by using radio-tracers. The concentration of dopamine in the brain can be observed by using a radio-tracer, 6-[18F]fluorodopa (FDOPA), with positron emission tomography (PET), and the dopamine kinetics can be described as compartmental models for tissues of the brain. The models for FDOPA kinetics are solved explicitly, but the solution shows a complicated form including several convolutions over time domain. Owing to the complicated form of the solution, graphical analyses such as Logan or Patlak analysis have been utilized as conventional methods over past decades. Because some kinetic constants for Parkinson's disease are estimated in the graphical analyses with the slope or intercept of the line obtained under various assumptions, only a limited set of parameters have approximately been estimated. We have analysed the compartmental models by using the Laplace transformation of differential equations and by algebraic computation with the aid of Gröbner base constructions. We have obtained a rigorous solution with respect to the kinetic constants over the Laplace domain. Here, we first derive a rigorous solution for the parameters, together with a discussion about the merits of the derivation. Next, we describe a procedure to determine the kinetic constants with the observed time-radioactivity curves. Last, we discuss the feasibility of our method, especially as a criterion for diagnosing Parkinson's disease.

Original languageEnglish
Title of host publicationAlgebraic Biology - Second International Conference, AB 2007, Proceedings
Pages110-124
Number of pages15
Publication statusPublished - Dec 1 2007
Event2nd International Conference on Algebraic Biology, AB 2007 - Castle of Hagenberg, Austria
Duration: Jul 2 2007Jul 4 2007

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume4545 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other2nd International Conference on Algebraic Biology, AB 2007
CountryAustria
CityCastle of Hagenberg
Period7/2/077/4/07

Fingerprint

Positron Emission Tomography
Parkinson's Disease
Positron emission tomography
Algebraic Approach
Positron-Emission Tomography
Parkinson Disease
Kinetics
Compartmental Model
Radio
Dopamine
Brain
Laplace Transformation
Radioactivity
Intercept
Laplace
Convolution
Time Domain
Slope
Differential equations
Tissue

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Yoshida, H., Nakagawa, K., Anai, H., & Horimoto, K. (2007). Exact parameter determination for Parkinson's disease diagnosis with PET using an algebraic approach. In Algebraic Biology - Second International Conference, AB 2007, Proceedings (pp. 110-124). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 4545 LNCS).

Exact parameter determination for Parkinson's disease diagnosis with PET using an algebraic approach. / Yoshida, Hiroshi; Nakagawa, Koji; Anai, Hirokazu; Horimoto, Katsuhisa.

Algebraic Biology - Second International Conference, AB 2007, Proceedings. 2007. p. 110-124 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 4545 LNCS).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Yoshida, H, Nakagawa, K, Anai, H & Horimoto, K 2007, Exact parameter determination for Parkinson's disease diagnosis with PET using an algebraic approach. in Algebraic Biology - Second International Conference, AB 2007, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 4545 LNCS, pp. 110-124, 2nd International Conference on Algebraic Biology, AB 2007, Castle of Hagenberg, Austria, 7/2/07.
Yoshida H, Nakagawa K, Anai H, Horimoto K. Exact parameter determination for Parkinson's disease diagnosis with PET using an algebraic approach. In Algebraic Biology - Second International Conference, AB 2007, Proceedings. 2007. p. 110-124. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
Yoshida, Hiroshi ; Nakagawa, Koji ; Anai, Hirokazu ; Horimoto, Katsuhisa. / Exact parameter determination for Parkinson's disease diagnosis with PET using an algebraic approach. Algebraic Biology - Second International Conference, AB 2007, Proceedings. 2007. pp. 110-124 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{5a8944eaac7e4a1e9d3724e1f73ac006,
title = "Exact parameter determination for Parkinson's disease diagnosis with PET using an algebraic approach",
abstract = "The mechanism of Parkinson's disease can be investigated at the molecular level by using radio-tracers. The concentration of dopamine in the brain can be observed by using a radio-tracer, 6-[18F]fluorodopa (FDOPA), with positron emission tomography (PET), and the dopamine kinetics can be described as compartmental models for tissues of the brain. The models for FDOPA kinetics are solved explicitly, but the solution shows a complicated form including several convolutions over time domain. Owing to the complicated form of the solution, graphical analyses such as Logan or Patlak analysis have been utilized as conventional methods over past decades. Because some kinetic constants for Parkinson's disease are estimated in the graphical analyses with the slope or intercept of the line obtained under various assumptions, only a limited set of parameters have approximately been estimated. We have analysed the compartmental models by using the Laplace transformation of differential equations and by algebraic computation with the aid of Gr{\"o}bner base constructions. We have obtained a rigorous solution with respect to the kinetic constants over the Laplace domain. Here, we first derive a rigorous solution for the parameters, together with a discussion about the merits of the derivation. Next, we describe a procedure to determine the kinetic constants with the observed time-radioactivity curves. Last, we discuss the feasibility of our method, especially as a criterion for diagnosing Parkinson's disease.",
author = "Hiroshi Yoshida and Koji Nakagawa and Hirokazu Anai and Katsuhisa Horimoto",
year = "2007",
month = "12",
day = "1",
language = "English",
isbn = "9783540734321",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
pages = "110--124",
booktitle = "Algebraic Biology - Second International Conference, AB 2007, Proceedings",

}

TY - GEN

T1 - Exact parameter determination for Parkinson's disease diagnosis with PET using an algebraic approach

AU - Yoshida, Hiroshi

AU - Nakagawa, Koji

AU - Anai, Hirokazu

AU - Horimoto, Katsuhisa

PY - 2007/12/1

Y1 - 2007/12/1

N2 - The mechanism of Parkinson's disease can be investigated at the molecular level by using radio-tracers. The concentration of dopamine in the brain can be observed by using a radio-tracer, 6-[18F]fluorodopa (FDOPA), with positron emission tomography (PET), and the dopamine kinetics can be described as compartmental models for tissues of the brain. The models for FDOPA kinetics are solved explicitly, but the solution shows a complicated form including several convolutions over time domain. Owing to the complicated form of the solution, graphical analyses such as Logan or Patlak analysis have been utilized as conventional methods over past decades. Because some kinetic constants for Parkinson's disease are estimated in the graphical analyses with the slope or intercept of the line obtained under various assumptions, only a limited set of parameters have approximately been estimated. We have analysed the compartmental models by using the Laplace transformation of differential equations and by algebraic computation with the aid of Gröbner base constructions. We have obtained a rigorous solution with respect to the kinetic constants over the Laplace domain. Here, we first derive a rigorous solution for the parameters, together with a discussion about the merits of the derivation. Next, we describe a procedure to determine the kinetic constants with the observed time-radioactivity curves. Last, we discuss the feasibility of our method, especially as a criterion for diagnosing Parkinson's disease.

AB - The mechanism of Parkinson's disease can be investigated at the molecular level by using radio-tracers. The concentration of dopamine in the brain can be observed by using a radio-tracer, 6-[18F]fluorodopa (FDOPA), with positron emission tomography (PET), and the dopamine kinetics can be described as compartmental models for tissues of the brain. The models for FDOPA kinetics are solved explicitly, but the solution shows a complicated form including several convolutions over time domain. Owing to the complicated form of the solution, graphical analyses such as Logan or Patlak analysis have been utilized as conventional methods over past decades. Because some kinetic constants for Parkinson's disease are estimated in the graphical analyses with the slope or intercept of the line obtained under various assumptions, only a limited set of parameters have approximately been estimated. We have analysed the compartmental models by using the Laplace transformation of differential equations and by algebraic computation with the aid of Gröbner base constructions. We have obtained a rigorous solution with respect to the kinetic constants over the Laplace domain. Here, we first derive a rigorous solution for the parameters, together with a discussion about the merits of the derivation. Next, we describe a procedure to determine the kinetic constants with the observed time-radioactivity curves. Last, we discuss the feasibility of our method, especially as a criterion for diagnosing Parkinson's disease.

UR - http://www.scopus.com/inward/record.url?scp=38149119014&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=38149119014&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9783540734321

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 110

EP - 124

BT - Algebraic Biology - Second International Conference, AB 2007, Proceedings

ER -