Abstract
Abstract: :
Purpose: To evaluate a hand–held infant keratometer (IK3) suitable for use with infants and young children that incorporates infrared imaging to minimize ambient illumination artifacts and improve the ability for the child to attend to the central visible fixation target, sufficient working distance to be non–threatening to infants and young children, and a telecentric lens to minimize variation in magnification that occurs with change in working distance. Methods: A single–ring (200 mm diameter) keratoscope using 12 infrared LEDs was attached to a prototype telecentric telephoto lens designed for use with a Sony TRV460 Digital–8 camcorder. The Sony camcorder is sensitive in the infrared and incorporates a 20:1 zoom lens, but does not allow focusing closer than 1 meter. For this reason, a telecentric telephoto lens was designed and constructed to allow a 25 mm field of view at a nominal distance of 200 mm. The telecentric design minimizes magnification variation for images within the depth of field, which in turn improves the accuracy of corneal curvature measurement. Two IR coated achromatic lenses of +50mm fl and –50mm fl were used, along with prototyping hardware from Edmund Industrial Optics (Barrington, NJ). The lens adds 70mm to the length of the camcorder, and serves to mount the ring keratoscope. The instrument was then used to measure astigmatism in a group of Native American schoolchildren known to have a high prevalence of corneal astigmatism. Results: Thirteen children averaging 85 months age (range 55–113 months) had corneal astigmatism measured by Retinomax K–Plus. Two of the children had less than 0.50D, four had from 0.5D to 1D, four from 1D to 2D, and 3 had more than 2D of corneal astigmatism (4.25 D max). At least three scorable keratoscope images could be obtained from all children. Custom image analysis software was written to fit an ellipse to the image data. A total of 71 images were scored. The median IK3 estimate of corneal astigmatism within each subject was then compared to the Retinomax estimate. The two measures of astigmatism agreed within 0.25D for 11 of 13 subjects (85%), and the greatest difference was 0.58D. The average difference (Rmax–IK3) was –0.034D with SD of 0.24D. No systematic bias was detected (t=0.51,p=0.61). Conclusions: The hand–held keratoscope allows useful measurement of corneal astigmatism from a test distance that is tolerated by infants and young children.
Keywords: astigmatism • refractive error development