Purchase this article with an account.
Howard C. Howland, Anupama D. Kumar; An Inexpensive Photorefractor for Screening for Myopia. Invest. Ophthalmol. Vis. Sci. 2012;53(14):3460.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To test the feasibility of using an inexpensive digital camera as an eccentric photorefractor to test for myopia.
The Canon PowerShot D10 digital camera has a linear flash bulb 25 mm distant from, and directly above, the edge of its camera lens, making it possible to photograph light from the double-pass linespread in the pupils of subjects. Eleven myopic and 3 emmetropic subjects’ right eyes (classified by the autorefractor of the Topcon KR-9000 Wavefront analyzer as having an equivalent refraction either of less than or greater than -0.5 D) were photographed at 2 meters distance to record the presence or absence of a myopic crescent in the horizontal and vertical meridians. In cases where a crescent was not initially seen, subjects were asked to wear positive lenses of increasing power to augment the degree of myopic defocus. If a crescent was seen, the magnitude of myopia was estimated as d = e/ (F*p*a) + 1/a - L, where e is the eccentricity of the flash (25 mm), F is the fraction of the pupil that is dark measured along the refracted meridian, p is the pupil diameter, a is the camera-to-subject distance and L is the power of the lens worn. The accuracy of this photorefractive test to detect myopia was compared to autorefractor measurements.
The mean meridional refraction for all subjects was -3.07 +/- 2.87 D. All 11 myopic subjects were correctly detected. One of the three emmetropes was falsely identified as a myope. In five of the 28 meridians measured we saw a crescent when it was not expected (assuming the meridional refraction of the autorefractor was correct).
We can conclude that the Canon PowerShot D10 can serve as a screening device for detecting myopia. If used with increasingly positive lenses in front of the subject’s eyes, its accuracy can be refined to the dioptric difference between the lens when a crescent is observed and the one used immediately before it when a crescent was not observed. Further testing of borderline myopes will allow us to plot the receiver operator curve.
This PDF is available to Subscribers Only