June 2020
Volume 61, Issue 7
Free
ARVO Annual Meeting Abstract  |   June 2020
Imaging Quantification of Inflammation – A Prospective Comparison of Clinicians to Automated Imaging Analysis in the Detection of Ocular Inflammation
Author Affiliations & Notes
  • Jordan D Deaner
    Uveitis, Cole Eye Institute || Cleveland Clinic, Cleveland, Ohio, United States
  • Arthi Venkat
    Uveitis, Cole Eye Institute || Cleveland Clinic, Cleveland, Ohio, United States
  • Kimberly Baynes
    Uveitis, Cole Eye Institute || Cleveland Clinic, Cleveland, Ohio, United States
  • Emily Fisher
    Uveitis, Cole Eye Institute || Cleveland Clinic, Cleveland, Ohio, United States
  • Jennifer Welsh
    Uveitis, Cole Eye Institute || Cleveland Clinic, Cleveland, Ohio, United States
  • Sumit Sharma
    Uveitis, Cole Eye Institute || Cleveland Clinic, Cleveland, Ohio, United States
  • Sunil K Srivastava
    Uveitis, Cole Eye Institute || Cleveland Clinic, Cleveland, Ohio, United States
  • Footnotes
    Commercial Relationships   Jordan Deaner, None; Arthi Venkat, None; Kimberly Baynes, None; Emily Fisher, None; Jennifer Welsh, None; Sumit Sharma, Alimera (C), Allergan (C), Bausch and Lomb (C), Clearside (C), Eyepoint (C), Genentech (C), Regeneron (C); Sunil Srivastava, Bausch and Lomb (C), Clearside (C), Eyepoint (C), Gilead (C), Novartis (C), Optos (C), Regeneron (C), Regenerxbio (C), Santen (C), Zeiss (C)
  • Footnotes
    Support  Unrestricted grant from Santen
Investigative Ophthalmology & Visual Science June 2020, Vol.61, 2076. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Jordan D Deaner, Arthi Venkat, Kimberly Baynes, Emily Fisher, Jennifer Welsh, Sumit Sharma, Sunil K Srivastava; Imaging Quantification of Inflammation – A Prospective Comparison of Clinicians to Automated Imaging Analysis in the Detection of Ocular Inflammation. Invest. Ophthalmol. Vis. Sci. 2020;61(7):2076.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose : To compare uveitis clinicians to a trained masked reader using automated imaging analysis in their ability to detect inflammatory disease activity

Methods : This is a prospective, observational, consecutive case series. Patients were examined and imaged at baseline and all follow-up visits over a 6-month period using ultra-wide field fluorescein angiography (UWFFA) and spectral domain optical coherence tomography (SD-OCT) of the anterior chamber and macula. Automated software was used to detect anterior chamber cell density (cells/mm3), retinal leakage index (percentage of retina leaking on UWFFA), and OCT segmentation of the macula. A uveitis specialist evaluated the patients by clinical exam alone. The clinician graded anterior chamber (AC) cell, vitreous cell (VC), and vitreous haze (VH) with a grade of greater than 1+ indicating inflammatory activity. In addition, the clinicians documented the presence of CME, active chorioretinal (CR) lesion(s), and active vasculitis. Based upon their overall exam, the physician determined if the patient was active. Independently, a trained masked reader then determined activity based solely upon the automated imaging results. The criteria for activity were AC cell density >8 cells/mm3, actual retinal subfield mean thickness > 350 microns, and total leakage index >4%.

Results : 50 eyes from 25 patients enrolled in the study and were evaluated at 81 visits to total 162 eye exams over the study period. Overall, masked readers were more likely to identify a patient as active (n=68), than clinicians (n=52) (X2, p<0.01). Clinicians determined activity by the presence of CME (n=14), VC (n=9), vasculitis (n=5), AC cell (n=3), VH (n=3), but never by active chorioretinal lesions (n=0). However, there were 32 patients who did not reach criteria for clinical activity that the physician still determined were active. Of these 32 examinations, 21 (71.9%) were deemed active by the masked reader. The masked reader determined activity by AC-OCT cell density (n=38), total leakage index (n=35), followed by OCT retinal subfield thickness (n=18).

Conclusions : Masked readers using automated imaging analysis are more likely to identify inflammatory activity than clinician examination alone. Interestingly, patients deemed active when clinical activity criteria were not met were usually detected by masked reader assessment.

This is a 2020 ARVO Annual Meeting abstract.

×
×

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.

×