April 2014
Volume 55, Issue 13
ARVO Annual Meeting Abstract  |   April 2014
Normative Databases for Imaging Instrumentation
Author Affiliations & Notes
  • Murray Fingeret
    Optometry, VA NY Health Care System, Hewlett, NY
  • Tony Realini
    Optometry, VA NY Health Care System, Hewlett, NY
  • John G Flanagan
    Optometry, VA NY Health Care System, Hewlett, NY
  • Paul H Artes
    Optometry, VA NY Health Care System, Hewlett, NY
  • Chris A Johnson
    Optometry, VA NY Health Care System, Hewlett, NY
  • Linda M Zangwill
    Optometry, VA NY Health Care System, Hewlett, NY
  • David F Garway-Heath
    Optometry, VA NY Health Care System, Hewlett, NY
  • Ian B Gaddie
    Optometry, VA NY Health Care System, Hewlett, NY
  • Vincent Michael Patella
    Optometry, VA NY Health Care System, Hewlett, NY
  • Footnotes
    Commercial Relationships Murray Fingeret, Canon (C), Carl Zeiss Meditec (C), Heidelberg Engineering (C), Topcon (C); Tony Realini, None; John Flanagan, Carl Zeiss Meditec (C), Heidelberg Engineering (C); Paul Artes, None; Chris Johnson, None; Linda Zangwill, None; David Garway-Heath, None; Ian Gaddie, None; Vincent Michael Patella, Carl Zeiss Meditec, inc. (E)
  • Footnotes
    Support None
Investigative Ophthalmology & Visual Science April 2014, Vol.55, 4759. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Murray Fingeret, Tony Realini, John G Flanagan, Paul H Artes, Chris A Johnson, Linda M Zangwill, David F Garway-Heath, Ian B Gaddie, Vincent Michael Patella; Normative Databases for Imaging Instrumentation. Invest. Ophthalmol. Vis. Sci. 2014;55(13):4759.

      Download citation file:

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

  • Supplements

Purpose: A symposium was held in July 2013 in which issues related to the development of imaging databases were discussed. A key lesson was that the development of reference databases (also called normative databases) for imaging platforms is limited by the combination of 1) a lack of standardization of methodology for developing such databases and 2) guidance from regulatory bodies to inform the design of such databases.

Methods: Automated optic nerve imaging are commonly used in glaucoma management. The ability to discriminate healthy from glaucomatous optic nerves, and to detect progression of glaucomatous optic neuropathy over time, is limited by the quality of reference databases associated with the existing commercial devices. In the absence of standardized rules governing the development of reference databases, each manufacturer's database differs in size, eligibility criteria, and ethnic make-up, among other key features.

Results: The cohort should be representative of the patients who will be tested using the instrument and should be drawn from the same population as the test subjects. Exclusion should be minimized to ensure that the database reflects the same profiles of co-morbidities that will be seen in the test subjects. The database should be large enough to characterize the reference population, including important covariates, with the limitation that developing reference databases is costly. With reference databases, we are interested in the tails of the distribution—the extreme values—and are asking whether a tested subject’s value is different from the average to be statistically unlikely. Some covariates, such as age, refractive error/axial length, race/ethnicity, and disc tilt, are known to affect optic disc imaging parameters. If the effect of such a covariate on the measurement is known and large enough to make a clinically-relevant difference, then stratification is justified. In regards to eligibility, a full visual field is the most important requirement to be included since optic nerve appearance should not be exclusion for an imaging normative database to reduce the possibility of developing a supranormal dataset.

Conclusions: Imaging databases may be improved by standardizing the eligibility requirements and modifying the design features. Consistency among devices in regards to how databases are developed is important if reliable evaluation among devices is to take place.

Keywords: 550 imaging/image analysis: clinical • 552 imaging methods (CT, FA, ICG, MRI, OCT, RTA, SLO, ultrasound) • 629 optic nerve  

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.