May 2006
Volume 47, Issue 13
Free
ARVO Annual Meeting Abstract  |   May 2006
Image Informatics Tools for the Analysis of Retinal Images
Author Affiliations & Notes
  • S.K. Fisher
    University of California Santa Barbara, Santa Barbara, CA
    Neuroscience Research Institute,
    Center for Bioimage Informatics,
  • J. Byun
    University of California Santa Barbara, Santa Barbara, CA
    Center for Bioimage Informatics,
    Dept. of Electrical and Computer Engineering,
  • D. Fedorov
    University of California Santa Barbara, Santa Barbara, CA
    Center for Bioimage Informatics,
    Dept. of Electrical and Computer Engineering,
  • N. Vu
    University of California Santa Barbara, Santa Barbara, CA
    Center for Bioimage Informatics,
    Dept. of Electrical and Computer Engineering,
  • B. Sumengen
    University of California Santa Barbara, Santa Barbara, CA
    Center for Bioimage Informatics,
    Dept. of Electrical and Computer Engineering,
  • M. Verardo
    University of California Santa Barbara, Santa Barbara, CA
    Neuroscience Research Institute,
    Center for Bioimage Informatics,
  • G.P. Lewis
    University of California Santa Barbara, Santa Barbara, CA
    Neuroscience Research Institute,
  • B.S. Manjunath
    University of California Santa Barbara, Santa Barbara, CA
    Center for Bioimage Informatics,
    Dept. of Electrical and Computer Engineering,
  • Footnotes
    Commercial Relationships  S.K. Fisher, None; J. Byun, None; D. Fedorov, None; N. Vu, None; B. Sumengen, None; M. Verardo, None; G.P. Lewis, None; B.S. Manjunath, None.
  • Footnotes
    Support  NIH grant EY00888, NSF grant ITR 0331697
Investigative Ophthalmology & Visual Science May 2006, Vol.47, 4225. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      S.K. Fisher, J. Byun, D. Fedorov, N. Vu, B. Sumengen, M. Verardo, G.P. Lewis, B.S. Manjunath; Image Informatics Tools for the Analysis of Retinal Images . Invest. Ophthalmol. Vis. Sci. 2006;47(13):4225.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose: : To develop software tools that can be used for the enhancement and quantitative analysis of retinal images, and to test these methods on a large retinal image database.

Methods: : We have assembled a database of over 2000 retinal images that was used for the development of multiple software tools for image analysis. This database is currently being used to study the responses of cell types in the retina before and after retinal detachment. The long–term objective is to provide new information processing techniques allowing the fully automated extraction of information from bio–molecular images.

Results: : The tools developed to manage and analyze retinal images on a large scale are currently available at UCSB's Center for Bioimage Informatics website (http://www.bioimage.ucsb.edu). These tools include: 1) a database that can be used for organizing, viewing and searching digital images, 2) software for a digital notebook that allows efficient editing, adding and uploading of metadata and digital images collected by any imaging technique, 3) a software method for local contrast enhancement that corrects unevenness created by illumination and different staining intensities within an image, 4) a software method for image segmentation which can be used to quantize protein distribution within regions of an image, 5) image mosaicing software that assembles large, seamless montages of overlapping images, 6) a detector for cell nuclei which was developed as an ImageJ plugin for accurately detecting and counting cell nuclei that performs with the same accuracy or better as manual counting and 7) an automated method for measuring retinal layer thickness and density to provide quantitative profiles in conjunction with the nuclear detector software.

Conclusions: : The retinal image database provides a means for querying raw images and their processed metadata, and combining the images with features for content–based retrieval. The tools developed during this project are useful for quantitatively analyzing large sets of biological images. Ultimately, these techniques will allow for the efficient, accurate and quantitative analysis of biological images as a means of understanding how cells in retina or other tissues respond to stress, injury, aging and disease.

Keywords: retinal detachment • image processing • imaging/image analysis: non-clinical 
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×