Likelihood inference for particle location in fluorescence microscopy

John Hughes, John Fricks, William Hancock

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

We introduce a procedure to automatically count and locate the fluorescent particles in a microscopy image. Our procedure employs an approximate likelihood estimator derived from a Poisson random field model for photon emission. Estimates of standard errors are generated for each image along with the parameter estimates, and the number of particles in the image is determined using an information criterion and likelihood ratio tests. Realistic simulations show that our procedure is robust and that it leads to accurate estimates, both of parameters and of standard errors. This approach improves on previous ad hoc least squares procedures by giving a more explicit stochastic model for certain fluorescence images and by employing a consistent framework for analysis.

Original languageEnglish (US)
Pages (from-to)830-848
Number of pages19
JournalAnnals of Applied Statistics
Volume4
Issue number2
DOIs
StatePublished - Jun 2010
Externally publishedYes

Fingerprint

Fluorescence Microscopy
Likelihood Inference
Fluorescence microscopy
Stochastic models
Standard error
Microscopic examination
Photons
Fluorescence
Estimate
Information Criterion
Likelihood Ratio Test
Microscopy
Random Field
Stochastic Model
Least Squares
Likelihood
Siméon Denis Poisson
Count
Photon
Estimator

Keywords

  • Fluorescence microscopy
  • Maximum likelihood methods
  • Molecular motor
  • Nanotechnology
  • Organelle
  • Particle tracking
  • Poisson random field

ASJC Scopus subject areas

  • Statistics, Probability and Uncertainty
  • Modeling and Simulation
  • Statistics and Probability

Cite this

Likelihood inference for particle location in fluorescence microscopy. / Hughes, John; Fricks, John; Hancock, William.

In: Annals of Applied Statistics, Vol. 4, No. 2, 06.2010, p. 830-848.

Research output: Contribution to journalArticle

Hughes, John ; Fricks, John ; Hancock, William. / Likelihood inference for particle location in fluorescence microscopy. In: Annals of Applied Statistics. 2010 ; Vol. 4, No. 2. pp. 830-848.
@article{09b8e7315853424c80bc0b4a07f9e562,
title = "Likelihood inference for particle location in fluorescence microscopy",
abstract = "We introduce a procedure to automatically count and locate the fluorescent particles in a microscopy image. Our procedure employs an approximate likelihood estimator derived from a Poisson random field model for photon emission. Estimates of standard errors are generated for each image along with the parameter estimates, and the number of particles in the image is determined using an information criterion and likelihood ratio tests. Realistic simulations show that our procedure is robust and that it leads to accurate estimates, both of parameters and of standard errors. This approach improves on previous ad hoc least squares procedures by giving a more explicit stochastic model for certain fluorescence images and by employing a consistent framework for analysis.",
keywords = "Fluorescence microscopy, Maximum likelihood methods, Molecular motor, Nanotechnology, Organelle, Particle tracking, Poisson random field",
author = "John Hughes and John Fricks and William Hancock",
year = "2010",
month = "6",
doi = "10.1214/09-AOAS299",
language = "English (US)",
volume = "4",
pages = "830--848",
journal = "Annals of Applied Statistics",
issn = "1932-6157",
publisher = "Institute of Mathematical Statistics",
number = "2",

}

TY - JOUR

T1 - Likelihood inference for particle location in fluorescence microscopy

AU - Hughes, John

AU - Fricks, John

AU - Hancock, William

PY - 2010/6

Y1 - 2010/6

N2 - We introduce a procedure to automatically count and locate the fluorescent particles in a microscopy image. Our procedure employs an approximate likelihood estimator derived from a Poisson random field model for photon emission. Estimates of standard errors are generated for each image along with the parameter estimates, and the number of particles in the image is determined using an information criterion and likelihood ratio tests. Realistic simulations show that our procedure is robust and that it leads to accurate estimates, both of parameters and of standard errors. This approach improves on previous ad hoc least squares procedures by giving a more explicit stochastic model for certain fluorescence images and by employing a consistent framework for analysis.

AB - We introduce a procedure to automatically count and locate the fluorescent particles in a microscopy image. Our procedure employs an approximate likelihood estimator derived from a Poisson random field model for photon emission. Estimates of standard errors are generated for each image along with the parameter estimates, and the number of particles in the image is determined using an information criterion and likelihood ratio tests. Realistic simulations show that our procedure is robust and that it leads to accurate estimates, both of parameters and of standard errors. This approach improves on previous ad hoc least squares procedures by giving a more explicit stochastic model for certain fluorescence images and by employing a consistent framework for analysis.

KW - Fluorescence microscopy

KW - Maximum likelihood methods

KW - Molecular motor

KW - Nanotechnology

KW - Organelle

KW - Particle tracking

KW - Poisson random field

UR - http://www.scopus.com/inward/record.url?scp=84869186750&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84869186750&partnerID=8YFLogxK

U2 - 10.1214/09-AOAS299

DO - 10.1214/09-AOAS299

M3 - Article

AN - SCOPUS:84869186750

VL - 4

SP - 830

EP - 848

JO - Annals of Applied Statistics

JF - Annals of Applied Statistics

SN - 1932-6157

IS - 2

ER -