June 2013
Volume 54, Issue 15
Free
ARVO Annual Meeting Abstract  |   June 2013
Development and Evaluation of an Objective Cataract Grading Metric
Author Affiliations & Notes
  • Dustin Morley
    R&D, LENSAR, Orlando, FL
  • Ajay Pillai
    R&D, LENSAR, Orlando, FL
  • Richard Olmstead
    R&D, LENSAR, Orlando, FL
  • Footnotes
    Commercial Relationships Dustin Morley, Lensar (E); Ajay Pillai, LENSAR, Inc. (E); Richard Olmstead, LensAR (E)
  • Footnotes
    Support None
Investigative Ophthalmology & Visual Science June 2013, Vol.54, 820. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Dustin Morley, Ajay Pillai, Richard Olmstead; Development and Evaluation of an Objective Cataract Grading Metric. Invest. Ophthalmol. Vis. Sci. 2013;54(15):820.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose: The purpose of this effort is to objectively quantify cataract density based on lens backscatter metrics derived from the LENSAR Augmented Reality (AR) imaging system and structure a diagnostic algorithm with a high correlation to clinical LOCS III cataract grade.

Methods: Between January 2012 and March 2012, 52 patients underwent Refractive Laser-assisted Cataract Surgery at the Asian Eye Institute (Makati City, Philippines). Each patient’s medical chart was retrospectively reviewed and data was recorded for LOCS III Cataract Grade. Cataracts were graded by a certified LOCS III grader. Six objective measures of lens backscatter were extracted from the AR Images taken at the time of cataract surgery. Nonlinear regression was used to optimize weights for each of the six measurements, which were then used to issue a lens scatter cataract grade (LSCG) based on a linear combination of the six weighted measures. The six measurements were Nuclear Opacity (NO), Anterior Opacity (AO), Posterior Opacity (PO), Nuclear Density (ND), Anterior Density (AD), and Posterior Density (PD). The resulting LSCG metric was then evaluated based on how closely it matched the existing LOCS III metric.

Results: Data from 38 patients were included in this analysis. Images from 14 patients were excluded for various reasons, most commonly due to poor image quality and/or the presence of image artifacts that cause problems for the automatic image processing algorithms, both of which frequently result from severe lens backscatter of exceedingly high grade cataracts. Mean NO, AO, and PO values were 0.01 ± 0.03, 0.01 ± 0.02, and 0.13 ± 0.10, respectively. Mean ND, AD, and PD values were 98.70 ± 24.78, 81.77 ± 12.94, and 125.59 ± 24.92, respectively. Average LOCS III Cataract Grade was 3.73 ± 1.20. Average LSCG was 3.69 ± 0.93. LSCG correlated directly with LOCS III cataract rank, with an absolute deviation of LSCG from LOCSIII of 0.67 ± 0.47 (Max = 1.6).

Conclusions: Automated AR image analysis shows promise for accurate and swift cataract severity grading. Initial results show a direct correlation when compared to LOCS III cataract rank by a certified grader. Further work is planned to increase the robustness of the algorithm and better address the effect of image artifacts on the algorithm.

Keywords: 445 cataract • 550 imaging/image analysis: clinical • 678 refractive surgery  
×
×

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.

×