• DocumentCode
    295171
  • Title

    Approximate maximum likelihood estimators for position and energy in scintillation cameras

  • Author

    Clinthorne, Neal H. ; Ng, Chor-yi ; Rogers, W. Leslie

  • Author_Institution
    Div. of Nucl. Med., Michigan Univ., Ann Arbor, MI, USA
  • Volume
    3
  • fYear
    1995
  • fDate
    9-12 May 1995
  • Firstpage
    1940
  • Abstract
    It is often assumed that the photon arrival process incident on the photodetector array in a scintillation camera is Poisson. Conditioned on the location and energy of the incident gamma-ray. A simple experiment, in which the covariance between the sensor outputs is empirically determined demonstrates that the observations are not independent conditioned merely on location and energy. We assume that this lack of independence arises from random parameters influencing the Poisson intensity or photodetector responses. Because, these responses can be difficult to either model or measure as a function of the nuisance parameters, we develop two simple approximations to the ideal maximum likelihood estimator for location and energy
  • Keywords
    approximation theory; cameras; gamma-ray detection; maximum likelihood estimation; optical signal detection; photodetectors; random processes; scintillation; stochastic processes; Poisson arrival process; Poisson intensity; approximate maximum likelihood estimators; covariance; energy estimation; experiment; incident gamma-ray; nuisance parameters; photodetector array; photodetector response; photon arrival process; position estimation; random parameters; scintillation cameras; sensor outputs; Biomedical engineering; Cameras; Covariance matrix; Extraterrestrial measurements; Maximum likelihood estimation; Nuclear medicine; Optoelectronic and photonic sensors; Photodetectors; Photonic crystals; Position measurement;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Acoustics, Speech, and Signal Processing, 1995. ICASSP-95., 1995 International Conference on
  • Conference_Location
    Detroit, MI
  • ISSN
    1520-6149
  • Print_ISBN
    0-7803-2431-5
  • Type

    conf

  • DOI
    10.1109/ICASSP.1995.480595
  • Filename
    480595