Toward objective selection of representative microscope images

Mia K. Markey, Michael Boland, Robert F. Murphy

Research output: Contribution to journalArticle

Abstract

Scientists wishing to communicate the essential characteristics of a pattern (such as an immunofluorescence distribution) currently must make a subjective choice of one or two images to publish. We therefore developed methods for objectively choosing a typical image from a set, with emphasis on images from cell biology. The methods involve calculation of numerical features to describe each image, calculation of similarity between images as a distance in feature space, and ranking of images by distance from the center of the feature distribution. Two types of features were explored image texture measures and Zernike polynomial moments, and various distance measures were utilized. Criteria for evaluating methods for assigning typicality were proposed and applied to sets of images containing more than one pattern. The results indicate the importance of using distance measures that are insensitive to the presence of outliers. For collections of images of the distributions of a lysosomal protein, a Golgi protein, and nuclear DNA, the images chosen as most typical were in good agreement with the conventional understanding of organelle morphologies. The methods described here have been implemented in a web server (http://murphylab.web.cmu.edu/services/TypIC).

Original languageEnglish (US)
Pages (from-to)2230-2237
Number of pages8
JournalBiophysical Journal
Volume76
Issue number4
StatePublished - 1999
Externally publishedYes

Fingerprint

Nuclear Proteins
Organelles
Fluorescent Antibody Technique
Cell Biology
DNA
lysosomal proteins

ASJC Scopus subject areas

  • Biophysics

Cite this

Toward objective selection of representative microscope images. / Markey, Mia K.; Boland, Michael; Murphy, Robert F.

In: Biophysical Journal, Vol. 76, No. 4, 1999, p. 2230-2237.

Research output: Contribution to journalArticle

Markey, MK, Boland, M & Murphy, RF 1999, 'Toward objective selection of representative microscope images', Biophysical Journal, vol. 76, no. 4, pp. 2230-2237.
Markey, Mia K. ; Boland, Michael ; Murphy, Robert F. / Toward objective selection of representative microscope images. In: Biophysical Journal. 1999 ; Vol. 76, No. 4. pp. 2230-2237.
@article{833b75d862c5405daef4b0a4056c90a4,
title = "Toward objective selection of representative microscope images",
abstract = "Scientists wishing to communicate the essential characteristics of a pattern (such as an immunofluorescence distribution) currently must make a subjective choice of one or two images to publish. We therefore developed methods for objectively choosing a typical image from a set, with emphasis on images from cell biology. The methods involve calculation of numerical features to describe each image, calculation of similarity between images as a distance in feature space, and ranking of images by distance from the center of the feature distribution. Two types of features were explored image texture measures and Zernike polynomial moments, and various distance measures were utilized. Criteria for evaluating methods for assigning typicality were proposed and applied to sets of images containing more than one pattern. The results indicate the importance of using distance measures that are insensitive to the presence of outliers. For collections of images of the distributions of a lysosomal protein, a Golgi protein, and nuclear DNA, the images chosen as most typical were in good agreement with the conventional understanding of organelle morphologies. The methods described here have been implemented in a web server (http://murphylab.web.cmu.edu/services/TypIC).",
author = "Markey, {Mia K.} and Michael Boland and Murphy, {Robert F.}",
year = "1999",
language = "English (US)",
volume = "76",
pages = "2230--2237",
journal = "Biophysical Journal",
issn = "0006-3495",
publisher = "Biophysical Society",
number = "4",

}

TY - JOUR

T1 - Toward objective selection of representative microscope images

AU - Markey, Mia K.

AU - Boland, Michael

AU - Murphy, Robert F.

PY - 1999

Y1 - 1999

N2 - Scientists wishing to communicate the essential characteristics of a pattern (such as an immunofluorescence distribution) currently must make a subjective choice of one or two images to publish. We therefore developed methods for objectively choosing a typical image from a set, with emphasis on images from cell biology. The methods involve calculation of numerical features to describe each image, calculation of similarity between images as a distance in feature space, and ranking of images by distance from the center of the feature distribution. Two types of features were explored image texture measures and Zernike polynomial moments, and various distance measures were utilized. Criteria for evaluating methods for assigning typicality were proposed and applied to sets of images containing more than one pattern. The results indicate the importance of using distance measures that are insensitive to the presence of outliers. For collections of images of the distributions of a lysosomal protein, a Golgi protein, and nuclear DNA, the images chosen as most typical were in good agreement with the conventional understanding of organelle morphologies. The methods described here have been implemented in a web server (http://murphylab.web.cmu.edu/services/TypIC).

AB - Scientists wishing to communicate the essential characteristics of a pattern (such as an immunofluorescence distribution) currently must make a subjective choice of one or two images to publish. We therefore developed methods for objectively choosing a typical image from a set, with emphasis on images from cell biology. The methods involve calculation of numerical features to describe each image, calculation of similarity between images as a distance in feature space, and ranking of images by distance from the center of the feature distribution. Two types of features were explored image texture measures and Zernike polynomial moments, and various distance measures were utilized. Criteria for evaluating methods for assigning typicality were proposed and applied to sets of images containing more than one pattern. The results indicate the importance of using distance measures that are insensitive to the presence of outliers. For collections of images of the distributions of a lysosomal protein, a Golgi protein, and nuclear DNA, the images chosen as most typical were in good agreement with the conventional understanding of organelle morphologies. The methods described here have been implemented in a web server (http://murphylab.web.cmu.edu/services/TypIC).

UR - http://www.scopus.com/inward/record.url?scp=0033065594&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0033065594&partnerID=8YFLogxK

M3 - Article

C2 - 10096918

AN - SCOPUS:0033065594

VL - 76

SP - 2230

EP - 2237

JO - Biophysical Journal

JF - Biophysical Journal

SN - 0006-3495

IS - 4

ER -