Comparison of computer-photoscreening with non-cycloplegic retinoscopy for amblyopiogenic risk factors in children

Chin Med J (Engl). 2000 Nov;113(11):1007-10.


Objective: To determine the sensitivity and specificity of using a computer-photoscreener and non-cycloplegic retinoscopy in the detection of amblyopiogenic factors in nine to fifty months old infants and children.

Methods: Three hundred children, nine to fifty months old, were screened with the computer-photoscreener and non-cycloplegic retinoscopy. With a blinded standardized clinical assessment as the standard, an overall comparison of the sensitivity of and specificity results obtained with the two techniques was made. Photoscreen images on the computer monitor screen were reviewed and analyzed immediately by two independent observers for indicators of amblyopiogenic risk factors. Simultaneously, the results were compared to the findings of a full ophthalmologic examination.

Results: The computer-photoscreener revealed a sensitivity of 94.6% and specificity of 90.1%, and the non-cycloplegic retinocopy revealed a sensitivity of 85.7% and specificity of 81.0% for the detection of amblyopiogenic risk factors, including hyperopia (+2.75 D or more), myopia (-1.50 D or more), astigmatism (1.75 D or more), anisometropia (2.00 D or more), ocular misalignment (5 degrees or more), and media opacity (1.5 mm or more).

Conclusions: The computer-photoscreener offers an opportunity to identify problems that limit vision, and could provide a feasible and sufficiently reliable screening technique in infants and preschool children to be screened successfully for amblyopiogenic risk factors.

Publication types

  • Comparative Study
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Amblyopia / diagnosis*
  • Child, Preschool
  • Diagnosis, Computer-Assisted*
  • Humans
  • Infant
  • Ophthalmoscopy / methods
  • Refraction, Ocular
  • Refractive Errors / diagnosis
  • Risk Factors
  • Sensitivity and Specificity
  • Vision Screening / instrumentation
  • Vision Screening / methods*