June 2015
Volume 56, Issue 7
Free
ARVO Annual Meeting Abstract  |   June 2015
Preliminary Evaluation of a Mobile App for Strabismus Screening
Author Affiliations & Notes
  • Shrinivas Pundlik
    Ophthalmology, Harvard Medical School, Schepens Eye Research Institute, Mass Eye and Ear, Boston, MA
  • Matteo Tomasi
    Ophthalmology, Harvard Medical School, Schepens Eye Research Institute, Mass Eye and Ear, Boston, MA
    EyePhone LLC, Cambridge, MA
  • Kevin Houston
    Ophthalmology, Harvard Medical School, Schepens Eye Research Institute, Mass Eye and Ear, Boston, MA
  • Gang Luo
    Ophthalmology, Harvard Medical School, Schepens Eye Research Institute, Mass Eye and Ear, Boston, MA
  • Footnotes
    Commercial Relationships Shrinivas Pundlik, None; Matteo Tomasi, EyePhone LLC (I); Kevin Houston, EyePhone LLC (I); Gang Luo, EyePhone LLC (I)
  • Footnotes
    Support None
Investigative Ophthalmology & Visual Science June 2015, Vol.56, 5212. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Shrinivas Pundlik, Matteo Tomasi, Kevin Houston, Gang Luo; Preliminary Evaluation of a Mobile App for Strabismus Screening. Invest. Ophthalmol. Vis. Sci. 2015;56(7 ):5212.

      Download citation file:


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

      ×
  • Supplements
Abstract
 
Purpose
 

An automated Hirschberg Test using photographic analysis of the corneal reflection has been suggested as a method to screen for strabismus. However, cost to develop and maintain a dedicated device can hinder accessibility. Smartphones offer a platform to dispense inexpensive and widely accessible eye-care solutions. We have developed a mobile app with the goals of obtaining rapid, convenient, and quantifiable strabismus measurements. We present the results of a preliminary evaluation experiment to determine the accuracy of the app.

 
Methods
 

When using the app, an examiner asks the patient to fixate at a particular point and simply points the phone camera to the patient's face. Real time eye detection is shown on the screen, after which a snapshot can be taken with the flash on. The app then localizes the iris center, the corneal reflection, and computes the eye deviation according to the offset of the corneal reflection from the iris center. To evaluate the accuracy of the app, four subjects without strabismus fixated binocularly at 11 points at various eccentricities from -5.8° to 5.8° (a total of about 20Δ range). Eye deviations with respect to the center point (0°) were computed for the left and right eyes separately, and compared with the ground truth. In a case study, one subject with known left exotropia was tested with the cover test and by the app. In this case, the app measured the difference in the deviations of the two eyes.

 
Results
 

The app measurements for the 8 eyes closely matched the ground truth, with a slight over estimation (line fitting slope=1.08, R2=0.99). The mean measurement error was 1.1 ± 0.9Δ.There was no difference between right and left eyes (p=0.18). In case of the patient with known strabismus, the cover test and the app found left exotropia of 24 ± 3.6Δ and 28.5 ± 3.5Δ, respectively.

 
Conclusions
 

Preliminary results indicate that the strabismus app can accurately measure a wide range eye deviations and further clinical testing of the app is warranted.  

 
Strabismus Measurement App. Before image capture the display shows the area of interest (green box) and detected eyes (red).
 
Strabismus Measurement App. Before image capture the display shows the area of interest (green box) and detected eyes (red).
 
 
Strabismus Measurement App. After image capture the display shows the iris detection, corneal reflection, and the computed eye deviation values.
 
Strabismus Measurement App. After image capture the display shows the iris detection, corneal reflection, and the computed eye deviation values.

 
×
×

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.

×