April 2014
Volume 55, Issue 13
Free
ARVO Annual Meeting Abstract  |   April 2014
EyeMark: Automated Image-based Biomarker Computation Tools for Monitoring Diabetic Retinopathy
Author Affiliations & Notes
  • Chaithanya Ramachandra
    Eyenuk LLC, Woodland hills, CA
  • Sandeep Bhat
    Eyenuk LLC, Woodland hills, CA
  • Muneeswar G Nittala
    Ophthalmology, USC Keck school of medicine, Los Angeles, CA
  • Srinivas R Sadda
    Ophthalmology, USC Keck school of medicine, Los Angeles, CA
  • Kaushal Solanki
    Eyenuk LLC, Woodland hills, CA
  • Footnotes
    Commercial Relationships Chaithanya Ramachandra, Eyenuk LLC (E); Sandeep Bhat, Eyenuk LLC (E); Muneeswar Nittala, Eyenuk LLC (F); Srinivas Sadda, Eyenuk LLC (F); Kaushal Solanki, Eyenuk LLC (E)
  • Footnotes
    Support None
Investigative Ophthalmology & Visual Science April 2014, Vol.55, 4827. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Chaithanya Ramachandra, Sandeep Bhat, Muneeswar G Nittala, Srinivas R Sadda, Kaushal Solanki; EyeMark: Automated Image-based Biomarker Computation Tools for Monitoring Diabetic Retinopathy. Invest. Ophthalmol. Vis. Sci. 2014;55(13):4827.

      Download citation file:


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

      ×
  • Supplements
Abstract
 
Purpose
 

Microaneurysm (MA) turnover rates (appearance and disappearance rate of MAs) are a good predictor of likelihood of progression to more severe retinopathy, establishing MA turnover as an excellent biomarker for diabetic retinopathy. Measuring this quantity from color fundus images involves two labor intensive steps: careful alignment of current and baseline images, and marking of individual MAs. This process is very time consuming and prone to error, if done entirely by human graders. We propose an automated MA turnover computation tool - EyeMark.

 
Methods
 

100 longitudinal retinal fundus color image sets with images in each set collected at least 6 months apart were obtained retrospectively from University of Southern California Keck School of Medicine. The MAs in these images were annotated by an expert grader. The longitudinal images were aligned using condition number theory based registration in the vesselness transform domain. Region of interest (ROI) was determined based on morphological filtering and Hessian matrix based computations to limit the pixels that are processed further. Using the ground truth annotation in a 50/50 train/test split, pixels were classified as MA/non-MA using novel multi-scale descriptors such as Gaussian derivatives and morphological filterbanks in a support vector machine framework. Classified pixels are grouped into blobs to determine the MAs in a given image. The MA locations in longitudinal images are matched to determine number of persistent MAs and to compute turnover rates.

 
Results
 

All longitudinal image pairs of fair/adequate quality were successfully registered even with multiple lesion changes, and different intensities between the image pair (Figure 1). MA detection has high sensitivity with AUROC of 0.92 at lesion/blob level. The turnover range (Figure 2) computed across longitudinal images was in good agreement with the turnover computed using the ground truth annotations.

 
Conclusions
 

Automated MA turnover computation requires two key challenges to be resolved - robust longitudinal image registration and accurate MA detection. With EyeMark we have tackled these issues and achieved accurate MA turnover computation.

 
 
Figure 1: Image alignment is robust to changing lesions.
 
Figure 1: Image alignment is robust to changing lesions.
 
 
Figure 2: Turnover rate ranges computed by Eyemark are in good agreement with groundtruth. A. Appeared MA turnover rate ranges. B. Disappeared MA turnover rate ranges.
 
Figure 2: Turnover rate ranges computed by Eyemark are in good agreement with groundtruth. A. Appeared MA turnover rate ranges. B. Disappeared MA turnover rate ranges.
 
Keywords: 550 imaging/image analysis: clinical • 499 diabetic retinopathy • 549 image processing  
×
×

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.

×