June 2023
Volume 64, Issue 8
Open Access
ARVO Annual Meeting Abstract  |   June 2023
Artificial intelligence characterization of central visual field patterns in distinct glaucomatous optic disc phenotypes
Author Affiliations & Notes
  • Kelvin Du
    Hamilton Glaucoma Center, Shiley Eye Institute, Viterbi Family Department of Ophthalmology, University of California San Diego, La Jolla, California, United States
  • Alireza Kamalipour
    Hamilton Glaucoma Center, Shiley Eye Institute, Viterbi Family Department of Ophthalmology, University of California San Diego, La Jolla, California, United States
  • Sasan Moghimi
    Hamilton Glaucoma Center, Shiley Eye Institute, Viterbi Family Department of Ophthalmology, University of California San Diego, La Jolla, California, United States
  • Pooya Khosravi
    University of California Irvine School of Medicine, Irvine, California, United States
  • Takashi Nishida
    Hamilton Glaucoma Center, Shiley Eye Institute, Viterbi Family Department of Ophthalmology, University of California San Diego, La Jolla, California, United States
  • Cristiana Vasile
    Hamilton Glaucoma Center, Shiley Eye Institute, Viterbi Family Department of Ophthalmology, University of California San Diego, La Jolla, California, United States
  • Kareem Latif
    Hamilton Glaucoma Center, Shiley Eye Institute, Viterbi Family Department of Ophthalmology, University of California San Diego, La Jolla, California, United States
  • Jeffrey M Liebmann
    Bernard and Shirlee Brown Glaucoma Research Laboratory, Department of Ophthalmology, Edward S. Harkness Eye Institute, Columbia University Irving Medical Center, New York, New York, United States
  • Massimo A Fazio
    Department of Ophthalmology and Vision Sciences, Heersink School of Medicine, The University of Alabama at Birmingham, Birmingham, Alabama, United States
  • Christopher A Girkin
    Department of Ophthalmology and Vision Sciences, Heersink School of Medicine, The University of Alabama at Birmingham, Birmingham, Alabama, United States
  • Christopher Bowd
    Hamilton Glaucoma Center, Shiley Eye Institute, Viterbi Family Department of Ophthalmology, University of California San Diego, La Jolla, California, United States
  • Linda M Zangwill
    Hamilton Glaucoma Center, Shiley Eye Institute, Viterbi Family Department of Ophthalmology, University of California San Diego, La Jolla, California, United States
  • Tara Javidi
    Department of Electrical and Computer Engineering, University of California San Diego, La Jolla, California, United States
  • Robert N Weinreb
    Hamilton Glaucoma Center, Shiley Eye Institute, Viterbi Family Department of Ophthalmology, University of California San Diego, La Jolla, California, United States
  • Footnotes
    Commercial Relationships   Kelvin Du None; Alireza Kamalipour Fight for Sight Foundation, Code F (Financial Support); Sasan Moghimi National Eye Institute, Code F (Financial Support), UC Tobacco-Related Disease Research Program , Code F (Financial Support); Pooya Khosravi None; Takashi Nishida Topcon, Code C (Consultant/Contractor); Cristiana Vasile None; Kareem Latif None; Jeffrey Liebmann None; Massimo Fazio National Eye Institute, Heidelberg Engineering GmbH, Topcon Corporation, Code F (Financial Support); Christopher Girkin National Eye Institute, Heidelberg Engineering GmbH, Topcon Corporation, Code F (Financial Support); Christopher Bowd None; Linda Zangwill Abbvie Inc., Digital Diagnostics, Code C (Consultant/Contractor), National Eye Institute, Carl Zeiss Meditec Inc., Heidelberg Engineering GmbH, Optovue Inc., Topcon Medical Systems Inc., Code F (Financial Support), Zeiss Meditec, Code P (Patent); Tara Javidi None; Robert Weinreb Aerie Pharmaceuticals, Allergan, Equinox, Eyenovia, Nicox, Topcon, Code C (Consultant/Contractor), Heidelberg Engineering, Carl Zeiss Meditec, Konan Medical, Optovue, Centervue, Bausch&Lomb, Topcon, Code F (Financial Support), Toromedes, Carl Zeiss Meditec, Topcon , Code P (Patent)
  • Footnotes
    Support  EY11008, EY19869, EY14267, EY027510, EY026574, P30EY022589, R21EY031125, EY023704, EY029058, EY034148, EY028284, BrightFocus Foundation (by the donors of the National Glaucoma Research Program, a program of the BrightFocus Foundation grant #2017122), Research to Prevent Blindness, UC Tobacco-Related Disease Research Program (T31IP1511).
Investigative Ophthalmology & Visual Science June 2023, Vol.64, 981. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Kelvin Du, Alireza Kamalipour, Sasan Moghimi, Pooya Khosravi, Takashi Nishida, Cristiana Vasile, Kareem Latif, Jeffrey M Liebmann, Massimo A Fazio, Christopher A Girkin, Christopher Bowd, Linda M Zangwill, Tara Javidi, Robert N Weinreb; Artificial intelligence characterization of central visual field patterns in distinct glaucomatous optic disc phenotypes. Invest. Ophthalmol. Vis. Sci. 2023;64(8):981.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose : To quantitatively characterize the patterns of 10-2 visual field (VF) damage in distinct glaucomatous optic disc phenotypes using an unsupervised artificial intelligence (AI) approach.

Methods : Archetypal analysis was applied to a total of 7885 10-2 VF tests of 1242 healthy subjects, glaucoma suspects, and glaucoma patients (2272 eyes) to identify the patterns of central VF damage. Optic disc phenotypes of 507 eyes (of 343 patients) with glaucoma diagnosis were categorized into generalized cup enlargement (GE), myopic (MY), focal ischemic (FI), and senile sclerotic (SS). The coefficients of 10-2 archetypal decompositions were averaged among the eyes of each specific phenotype.

Results : Sixteen distinct 10-2 VF archetypes (ATs) were identified. Among the participants with available optic disc phenotypes information, 177 (51.6%) were female, and 151 (44.0%) were of African descent. The mean age, 24-2 VF mean deviation (MD), and 10-2 VF MD were 72.3 years (71.3, 73.3), -8.57 dB (-9.25, -7.88), and -7.69 dB (-7.02, -8.36), respectively. Senile sclerotic was the most common phenotype (n = 164, 32.3%), followed by GE (n = 152, 30.0%), FI (n = 118, 23.3%), and MY (n = 73, 14.4%). Even though the intact AT (AT3) was the most frequent in all phenotypes, its average coefficient was lower in the SS (27.1%) and FI (28.6%) compared to the GE (51.2%) and MY (37.6%) phenotypes. The next most frequent patterns of central VF damage differed by each phenotype (Figure 1). The most frequent patterns of central VF damage of the FI and GE phenotypes were limited to a single hemifield. The total loss AT (AT13) coefficient was higher in the SS (6.4%) and MY (5.9%) compared to the FI (2.1%) and GE (3.6%) phenotypes.

Conclusions : The extent and characteristics of central VF damage vary by the distinct appearance of optic disc phenotypes in glaucoma patients. The overall severity of central VF damage was higher in the SS and FI phenotypes, and the FI phenotype was associated with more localized patterns of central VF loss. Precise quantification of central VF damage and their relevance to optic disc phenotypes by AI methods may be useful for optimal resource reallocation and guiding the management of glaucoma patients.

This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.

 

Figure 1. Representative examples of optic disc phenotypes, along with their four most common 10-2 visual field archetypes (AT) and their respective average coefficients.

Figure 1. Representative examples of optic disc phenotypes, along with their four most common 10-2 visual field archetypes (AT) and their respective average coefficients.

×
×

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.

×