July 2019
Volume 60, Issue 9
Open Access
ARVO Annual Meeting Abstract  |   July 2019
The Spatial Frequency Content of Urban and Indoor Environments as a Potential Risk Factor for Myopia Development.
Author Affiliations & Notes
  • Daniel Ian Flitcroft
    Ophthalmology, Childrens University Hospital, Dublin, Ireland
    Ophthalmology, University College Dublin, Dublin, Ireland
  • Elise N Harb
    Optometry, University of California Berkeley, California, United States
  • Christine Frances Wildsoet
    Optometry, University of California Berkeley, California, United States
  • Footnotes
    Commercial Relationships   Daniel Flitcroft, None; Elise Harb, None; Christine Wildsoet, None
  • Footnotes
    Support  None
Investigative Ophthalmology & Visual Science July 2019, Vol.60, 6452. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Daniel Ian Flitcroft, Elise N Harb, Christine Frances Wildsoet; The Spatial Frequency Content of Urban and Indoor Environments as a Potential Risk Factor for Myopia Development.. Invest. Ophthalmol. Vis. Sci. 2019;60(9):6452.

      Download citation file:

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

  • Supplements

Purpose : To examine the hypothesis that the spatial frequency spectra of urban and indoor environments differ from the natural environment and have features that resemble defocussed images, thereby potentially creating a stimulus for eye growth and myopia.

Methods : 191 images were collected to represent a range of environments, i.e. natural, mixed and man-made (urban) outdoors, and domestic, student accommodation and university building indoors. Images were processed in Matlab (Mathworks Inc) to map the camera color model to human cone sensitivities and create photopic luminance images. These images were then analyzed to generate rotationally averaged spatial frequency spectra, and the slope of the log amplitude vs log spatial frequency relationship (SF slope) calculated for each image. Scene luminance in cd/m2 was estimated for each image from camera exposure and aperture settings.

Results : The natural images conformed to the well-recognised 1/f relationship in terms of spatial frequency with an SF slope of -1.06 (S.E.: 0.015). The images from outdoor urban and indoor environments all had significantly steeper slopes than the natural outdoors image set (Figs. 1 & 2): mixed outdoors, -1.17 (S.E: 0.024), urban outdoors, -1.26 (s.e.0.021), student accommodation indoors, -1.34 (S.E: 0.030), domestic indoors, -1.40 (s.e. 0.077) and university building indoors, -1.51 (S.E: 0.039). Imposing optical defocus on a natural image scene over a range of -4 to +4 D resulted in a symmetrical change in the slope of the frequency spectrum of -0.297/D (r2=0.949,95% CI -0.359 -0.235).[EH1] The changes in slope for images from different man-made environments corresponded to defocus levels of 0.66 D, outdoor urban, 1.12 D, indoors and 1.51 D, inside university building indoors. Bangerter foils of a level that produce myopia in animal models (i.e. 0.4) had a similar effect, increasing the SF slope to -1.53. Across all images there was also a significant relationship between estimated log luminance and SF slope (r2=0.25,95%CI 0.039 0.065).

Conclusions : Compared to natural outdoor environments, urban outdoor and indoor environments have spatial frequency characteristics that are similar to slightly defocussed (blurred) images, equivalent to 0.66 to 1.51 D. These findings may help to explain the apparent myopigenic nature of the urban and indoor environments.

This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.




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.