Computer Aided Detection Of Strabismus In Humans Using Computer Vision Techniques

Abstract

Strabismus, also known as squint, is a common eye condition in which the eyes do not properly align with each other. This can cause double vision, blurred vision, and difficulty with depth perception. Early diagnosis and treatment of strabismus are important to prevent amblyopia, a condition in which the vision in one eye does not develop properly. According to studies, 2-4% of Indians suffer from strabismus, which most commonly affects children. The impact of strabismus on human life is severe. Strabismus patients experience social and psychological concerns in addition to visual impairments. Children who have strabismus may experience difficulties with their social interactions, academic success, self-assurance, and confidence. People generally have a more negative view of individuals with strabismus compared to those with normal eye alignment. Research on adults has shown that having a significant horizontal strabismus can make it harder to find employment. Interestingly, it appears that women's employability is more affected by this condition than men's. Computer vision-based strabismus detection systems have a number of advantages over traditional methods of diagnosis. They are non-invasive, objective, and can be used to screen large populations for strabismus. Additionally, computer vision systems can be used to monitor the progression of strabismus over time and to assess the response to treatment. The proposed method uses a front facing portrait image of a patient to measure the deviation in positional similarity between their two eyes. Our system achieves an accuracy of 96%. We have used a dataset of 1000 images, consisting of eyes both with and without strabismus. This aims to provide ophthalmologists with clear data for diagnosing strabismus, making strabismus screening more accessible for individuals in remote areas, and assisting ophthalmologists in diagnosing strabismus more quickly and cost-effectively.