A computerized image analysis method for measuring elastic tissue

Thomas J. Flotte, Johanna M. Seddon, Yuqing Zhang, Robert J. Glynn, Kathleen M. Egan, Evangelos S. Gragoudas

Research output: Contribution to journalArticle

29 Scopus citations

Abstract

A computerized image analysis technique for quantifying area fractions of elastic tissue in light microscopic sections is presented that considers the nonuniform distribution of the elastic tissue. The method uses Verhoeff-van Gieson stained sections. The analysis is stratified vertically in three uniform layers starting at the dermo-epidermal junction and horizontally by two schemes. A standard method consists of five equally spaced measurements. The densest method uses the three areas that contain the most elastic tissue. The area fractions are determined by counting the positive and total pixels (thresholding). A validity test utilizing independent physical measurements demonstrated differences of no more than 1.7%. Reliability tests for reading the same section on different days and adjacent sections showed no significant differences (p < 0.05) between the readings. Reliability tests of sections using different stain lots and adjacent biopsy sites also did not have significant differences. This method may be particularly useful for studies in which the distribution of the material to be measured may be very uneven, such as in solar elastosis.

Original languageEnglish (US)
Pages (from-to)358-362
Number of pages5
JournalJournal of Investigative Dermatology
Volume93
Issue number3
DOIs
StatePublished - Sep 1989

ASJC Scopus subject areas

  • Biochemistry
  • Molecular Biology
  • Dermatology
  • Cell Biology

Fingerprint Dive into the research topics of 'A computerized image analysis method for measuring elastic tissue'. Together they form a unique fingerprint.

  • Cite this