Development of a computer system for strabismus screening

The Strabismus affects about 4% of the children population, causing amblyopia disease. However, this problem can be found by the Hirschberg test that is one of the exams used to detect this pathology. This work presents a methodology for automatic detection of strabismus in digital images through the Central corneal light reflex ratio (CCLRR) based on digital image processing. The proposed algorithm was organized into three stages: (1) skin detection was analyzed using the threshold of RGB color model (2) the locations of the eyes were detected using the morphological operation technique, position, threshold of gray level and shape. (3) The detection of the corneal light reflex were automatically detected by the grey levels in the range from histogram equation and locating of limbus. The methodology has produced results on the range 94.17% of accuracy, 97.23% of sensitivity and 73.08% of specificity.