Investigative Ophthalmology & Visual Science Cover Image for Volume 62, Issue 8
June 2021
Volume 62, Issue 8
Open Access
ARVO Annual Meeting Abstract  |   June 2021
Objective method for limbus detection of anterior segment topography
Author Affiliations & Notes
  • Abhishek Bhat
    Mojo Vision, California, United States
  • Kenneth Duy Tran
    Mojo Vision, California, United States
  • Christine W Sindt
    Department of Ophthalmology, University of Iowa, Iowa, United States
  • Klara Barbarossa
    Mojo Vision, California, United States
  • Megan Bauer
    Department of Ophthalmology, University of Iowa, Iowa, United States
  • Raymond Lum
    Mojo Vision, California, United States
  • Kuangmon Ashley Tuan
    Mojo Vision, California, United States
  • Footnotes
    Commercial Relationships   Abhishek Bhat, None; Kenneth Tran, None; Christine Sindt, EyePrint Prosthetics LLC (P); Klara Barbarossa, None; Megan Bauer, None; Raymond Lum, None; Kuangmon Tuan, None
  • Footnotes
    Support  None
Investigative Ophthalmology & Visual Science June 2021, Vol.62, 665. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Abhishek Bhat, Kenneth Duy Tran, Christine W Sindt, Klara Barbarossa, Megan Bauer, Raymond Lum, Kuangmon Ashley Tuan; Objective method for limbus detection of anterior segment topography. Invest. Ophthalmol. Vis. Sci. 2021;62(8):665.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose : Successful contact lens designs are predicated on the accurate identification of the limbus. Custom scleral lens designs can now be fabricated using 3D ocular surface maps from scanning and impression-based methods. Utilizing 3D topography maps, an objective method of defining the limbus was developed.

Methods : Thirty impression-based 3D topography maps were selected based on ocular conditions that do not affect the limbal region.

Subjective Detection: An expert clinician identified at least 5 points on the topography map using the EyePrint Designer (EPD) software to represent the limbal margin. Visual selection of points was guided by known anatomical features such as expected curvature change and limbus size.

Objective Detection: A custom algorithm in MATLAB detects changes in 3D ocular surface profile at the limbal region. An initial estimate of candidate limbal points represent locations with large gradient magnitudes. An adaptive varying threshold is applied since the gradient magnitude of the limbus is not constant across all meridians. Candidate points away from the probable limbal region are rejected. Further testing for outliers using Random Sample Consensus (RANSAC) is done to remove points that do not fit an elliptical profile. The detected limbal points tend not to be planar, resulting in a best fit hyperbolic paraboloid shape.

Comparison: Limbal parameters from subjective and objective methods collected using the EPD software were compared. All objective limbus designs were reviewed by another expert clinician. Lens designs with >75% similarity of the limbal region between the two methods were clinically acceptable.

Results : Limbal parameters were reported as: major diameter, minor diameter, and axis of major diameter. The average difference between the two methods were 0.16mm, 0.07mm, and 1.92 degrees, respectively. The median difference between the two methods were 0.05mm, 0.03mm, and 2.95 degrees, respectively. Review of objectively determined limbal points found 80% of designs to be clinically acceptable.

Conclusions : This objective method of identifying the limbus may be an effective way of guiding novice contact lens designers to create custom scleral lenses. Further investigation of irregular ocular surfaces is necessary to address the broader ocular demographic who wear scleral lenses.

This is a 2021 ARVO Annual Meeting abstract.

 

Comparison of limbus detection methods

Comparison of limbus detection methods

 

Candidate and selected limbal points on surface gradient profile

Candidate and selected limbal points on surface gradient profile

×
×

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.

×