Prueba de detección para la detección temprana de miopía basada en imágenes híbridas

Myopia screening is crucial as early intervention can slow its progression reducing the risk of vision-threatening diseases. We present an innovative method that can be used by non-experts, such as parents. The method relies on "hybrid images", an optical illusion merging two images with different spatial frequencies, creating a perception that changes depending on visual acuity. Hybrid images are created by combining one image with reduced low spatial frequencies and another with attenuated high spatial frequencies. Various cutoff frequencies have been tested, and an optimization phase was run to select the best-performing images using subjects with known visual acuity altered by positive lenses. We used animal images, as they are well-suited for hybrid images due to their facial features and help frame the screening as a game, an essential aspect for children and engaging for young people. The optimized hybrid images were tested monocularly on 81 young adults (mean age [Formula: see text]) with and without visual correction. The method showed high accuracy in distinguishing visual acuity above or below +0.1 logMar, with sensitivity ranging from 0.86 to 0.95 and specificity from 0.94 to 0.97. Presented as a proof of concept, the method demonstrates high specificity, highlighting its potential for non-clinical screenings (e.g., schools) using only five printed images.