Investigative Ophthalmology & Visual Science Cover Image for Volume 60, Issue 9
July 2019
Volume 60, Issue 9
Open Access
ARVO Annual Meeting Abstract  |   July 2019
Visual outcomes predicted by macular morphology of patients with neovascular age-related macular degeneration or polypoidal choroidal vasculopathy using an automated segmentation algorithm
Author Affiliations & Notes
  • MINSU JANG
    Ophthalmology, Konkuk university medical center, Seoul, Korea (the Republic of)
  • Hyungwoo Lee
    Ophthalmology, Konkuk university medical center, Seoul, Korea (the Republic of)
  • Hyung Chan Kim
    Ophthalmology, Konkuk university medical center, Seoul, Korea (the Republic of)
  • Hyewon Chung
    Ophthalmology, Konkuk university medical center, Seoul, Korea (the Republic of)
  • Footnotes
    Commercial Relationships   MINSU JANG, None; Hyungwoo Lee, None; Hyung Chan Kim, None; Hyewon Chung, None
  • Footnotes
    Support   National Research Foundation of Korea (NRF) and funded by the Ministry of Science and ICT (NRF-2017R1E1A1A01073964)
Investigative Ophthalmology & Visual Science July 2019, Vol.60, 1359. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      MINSU JANG, Hyungwoo Lee, Hyung Chan Kim, Hyewon Chung; Visual outcomes predicted by macular morphology of patients with neovascular age-related macular degeneration or polypoidal choroidal vasculopathy using an automated segmentation algorithm. Invest. Ophthalmol. Vis. Sci. 2019;60(9):1359.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose : To investigate factors predictive of visual outcomes in patients with neovascular age-related macular degeneration (nAMD) or polypoidal choroidal vasculopathy (PCV) treated with anti-vascular endothelial growth factor (anti-VEGF) from clinical and spectral domain optical coherence tomography (SD-OCT) parameters using an automated segmentation algorithm based on a convolutional neural network (CNN).

Methods : Quantification of intraretinal fluid (IRF), subretinal fluid (SRF), pigment epithelial detachment (PED), and subretinal hyperreflective material (SHRM) from SD-OCT images in patients with nAMD or PCV was achieved by the CNN based on automated segmentation algorithm established in our previous work. A total of 181 eyes including, 133 nAMD and 48 PCV, were evaluated. The following parameters were used to examine the correlations between baseline anatomic characteristics on SD-OCT: best-corrected visual acuity (BCVA) at baseline and at 12 months; and the total number of anti-VEGF injections.

Results : Larger gains in the BCVA at 12 months were correlated with poorer BCVA and a smaller quantity of SHRM at 12 months in eyes with nAMD (P<0.0005, P<0.012, respectively); larger gains in the BCVA were correlated with only poorer BCVA at baseline in eyes with PCV (P<0.0005). The area of SRF and PED at baseline was larger in eyes with PCV than those with nAMD. No significant difference in the BCVA at 12 months between eyes with nAMD and PCV were found. However, the total number of injections was higher in eyes with PCV than that of eyes with nAMD (P=0.006).

Conclusions : Poorer BCVA at baseline predicted larger gains in BCVA in both subtypes treated with anti-VEGF. Also, eyes with nAMD had greater improvement of BCVA if lesser SHRM was present at 12 months. Eyes with PCV achieved visual gain comparable to eyes with nAMD despite the larger quantity of SRF and PED at baseline and large number of total injections than eyes with nAMD.

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.

×