June 2023
Volume 64, Issue 9
Open Access
ARVO Imaging in the Eye Conference Abstract  |   June 2023
Robot-Assisted Glaucoma Assessment Based on Visual Fields
Author Affiliations & Notes
  • Swapneel Mandal
    Collierville High School, Collierville, Tennessee, United States
  • Xiaoqin Huang
    The University of Tennessee Health Science Center, Memphis, Tennessee, United States
  • Yeganeh Madadi
    The University of Tennessee Health Science Center, Memphis, Tennessee, United States
  • Hina Raja
    The University of Tennessee Health Science Center, Memphis, Tennessee, United States
  • Asma Poursoroush
    The University of Tennessee Health Science Center, Memphis, Tennessee, United States
  • Mohammad Delsoz
    The University of Tennessee Health Science Center, Memphis, Tennessee, United States
  • Siamak Yousefi
    The University of Tennessee Health Science Center, Memphis, Tennessee, United States
  • Footnotes
    Commercial Relationships   Swapneel Mandal, None; Xiaoqin Huang, None; Yeganeh Madadi, None; Hina Raja, None; Asma Poursoroush, None; Mohammad Delsoz, None; Siamak Yousefi, Remidio (R)
  • Footnotes
    Support  NIH Grant EY031725, EY033005; Bright focus Foundation; Research to Prevent Blindness (RPB)
Investigative Ophthalmology & Visual Science June 2023, Vol.64, PB0022. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Swapneel Mandal, Xiaoqin Huang, Yeganeh Madadi, Hina Raja, Asma Poursoroush, Mohammad Delsoz, Siamak Yousefi; Robot-Assisted Glaucoma Assessment Based on Visual Fields. Invest. Ophthalmol. Vis. Sci. 2023;64(9):PB0022.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose : To employ Alborz, a humanoid robot, to objectively assess visual fields and predict glaucoma progression.

Methods : We have developed a deep archetypal analysis (DAA) model to identify different patterns of visual field (VF) loss from 2,231 abnormal VFs collected from 205 eyes of 176 participants in the Ocular Hypertensive Treatment Study (OHTS) (Yousefi et al., Ophthalmology, 2022). We then investigated whether DAA-identified patterns of visual field loss can be used to predict rapid glaucoma progression. The DAA model was then integrated into Alborz, an artificial intelligence-enabled humanoid robot from the Nao family with natural language processing (NLP) capabilities and the ability to communicate findings. Alborz can apply the DAA model to provide a more objective glaucoma assessment, identify patients who may undergo rapid glaucoma progression, and provide direct communication of findings.

Results : The average mean deviation (MD) was -2.7 dB (SD: 2.4) in vision loss from the initial assessment of all eyes. Based on data from subsequent visits, a total of 50 of the 205 eyes were determined to be progressing rapidly, with MD rates worse than -1 dB/year. A total of 18 distinct spatial patterns of VF loss were identified, including temporal wedge, partial arcuate, nasal step, and paracentral retinal regions. Using the integrated DAA model, Alborz can identify patterns of visual field defects and predict future rapid glaucoma progression based on the weights of DAA patterns.

Conclusions : An unsupervised machine learning model integrated into Alborz, a humanoid robot, can decompose visual fields into DAA patterns of visual field loss which, in turn, can be used to identify patients likely to experience rapid glaucoma progression. Such objective glaucoma assessment models show promising potential for assisting clinicians to make more unbiased and informed decisions and to monitor glaucoma progression, in general, more accurately. In addition, the ability of Alborz to process data objectively and simultaneously use natural language processing to communicate findings creates an important opportunity to enhance a smoother integration of artificial intelligence systems into glaucoma clinical practice.

This abstract was presented at the 2023 ARVO Imaging in the Eye Conference, held in New Orleans, LA, April 21-22, 2023.

 

Figure 1. Diagram of the proposed robot-assisted model for glaucoma assessment.

Figure 1. Diagram of the proposed robot-assisted model for glaucoma assessment.

×
×

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.

×