Artificial intelligence detection of refractive eye diseases using certainty factor and image processing

Abstract

Refractive errors are defined as an impairment in the eye’s capacity to focus light, resulting in the formation of blurred or unfocused images. These issues arise from alterations in the shape of the cornea, the length of the eyeball, or the aging of the crystalline lens. It is anticipated that the prevalence of visual impairment will increase in conjunction with global population growth. At present, a significant number of countries have not yet accorded sufficient priority to eye health within their healthcare systems. This has resulted in insufficient awareness and reluctance to seek costly specialized care. This study proposes the development of an advanced refractive eye disease detection system with the objective of improving diagnostic accuracy, disseminating disease information, and reducing financial barriers to specialist consultation. The research employs certainty factor (CF) methods and image processing with feature extraction. The initial results demonstrate the potential for identifying specific refractive eye diseases with high certainty through the analysis of symptoms and the examination of photographs of the eye. The proposed approach provides an alternative method for diagnosing refractive eye diseases, which could enhance access to refractive eye care services and reduce the economic burden on patients.