July 2018
Volume 59, Issue 9
Open Access
ARVO Annual Meeting Abstract  |   July 2018
Beyond the average threshold: Alternatives in the analysis of microperimetry data
Author Affiliations & Notes
  • Iain Robert Wilson
    Nuffield Laboratory of Ophthalmology, University of Oxford, Oxford, United Kingdom
  • Jasleen Jolly
    Nuffield Laboratory of Ophthalmology, University of Oxford, Oxford, United Kingdom
    Oxford Eye Hospital, Oxford University Hospitals NHS Foundation Trust, Oxford, Oxfordshire, United Kingdom
  • Susan Downes
    Nuffield Laboratory of Ophthalmology, University of Oxford, Oxford, United Kingdom
    Oxford Eye Hospital, Oxford University Hospitals NHS Foundation Trust, Oxford, Oxfordshire, United Kingdom
  • Robert E MacLaren
    Nuffield Laboratory of Ophthalmology, University of Oxford, Oxford, United Kingdom
    Oxford Eye Hospital, Oxford University Hospitals NHS Foundation Trust, Oxford, Oxfordshire, United Kingdom
  • Footnotes
    Commercial Relationships   Iain Wilson, None; Jasleen Jolly, None; Susan Downes, None; Robert MacLaren, None
  • Footnotes
    Support  NIHR - i4i
Investigative Ophthalmology & Visual Science July 2018, Vol.59, 1698. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Iain Robert Wilson, Jasleen Jolly, Susan Downes, Robert E MacLaren; Beyond the average threshold: Alternatives in the analysis of microperimetry data. Invest. Ophthalmol. Vis. Sci. 2018;59(9):1698.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose : Microperimetry is being used increasingly in interventional trials, such as for gene therapy. Comprehensive analysis of the results incorporating spatial information can be complicated, resulting in the average threshold being used in most cases. This can easily be skewed by multiple non-seeing points so treatment effects are lost in the process. The aim of this study is to evaluate a new analysis technique for the results of microperimetry, and to determine if fundus imaging factors impact test-retest variability.

Methods : We developed and evaluated a new technique (based on sum of absolute differences) that we call thresholded sum of differences (TSD). The TSD technique, which incorporates point wise test-retest variability, generates a single value and can be used for comparing multiple measurements.
Evaluation was performed first with simulated data; a central scotoma was generated from a 2D Gaussian curve, with varying in depth and radius. The sensitivity to change of the average threshold and TSD to the simulation was compared. Secondly, we evaluated how the techniques correlate with retest variability, as well as each other, using real data from a variety of patient groups.
We further investigated factors leading to test retest variability; standard reports were used along with data generated from variations in the fundus images captured in repeated sessions; blur, brightness and registration error. Repeated MAIA microperimetry measures were collected from 19 normal, 30 glaucoma and 24 choroideremia subjects.

Results : Simulation: TSD detected 92% of the simulated scotomas compared to 71% for average threshold, using test-retest values of 4.3dB and 2.0dB respectively.
Real Data: The point wise test-retest variability of normal, glaucoma and choroideremia were; 4.3dB, 8.16dB and 9.45dB. Analysis revealed TSD correlates with average threshold(p<0.001 r=-0.99). Test retest variability correlated with TSD (p<0.001, r=0.85), and average threshold (p<0.001, r=-0.62). Blur(p=0.16), brightness(p=0.94), registration error(p=0.65), fixation errors(p=0.08) and disease(p=0.79) did not contribute to test-retest variability.

Conclusions : Thresholded sum of differences is a novel analysis technique for microperimetry data, allowing the incorporation of spatial data to increase sensitivity to local changes. The technique can be employed for single clinical measurements or as an outcome measure for longitudinal trials.

This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.

×
×

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.

×