EXPLORATION OF AN ANDROID-BASED MYOPIA DETECTION AND PREVENTION SYSTEM IN CHILDREN USING THE EXTREME LEARNING MACHINE METHOD WITH A RISK FACTOR ANALYSIS AND EYE EXERCISE THERAPY APPROACH

Abstract

Myopia, or nearsightedness, is a visual impairment whose prevalence continues to increase, particularly among children. This condition is influenced by both genetic and environmental factors, such as excessive use of digital devices and insufficient outdoor activities. These challenges highlight the need for solutions that not only enable early detection but also provide effective preventive measures. This study aims to explore an Android-based system for the detection and prevention of myopia in children using the Extreme Learning Machine (ELM) method, combined with a risk factor analysis approach and eye exercise therapy. The research employs a qualitative method with a library research approach, utilizing various relevant scientific sources, including journals, books, and conference proceedings. Data were collected through systematic searches in academic databases using keywords aligned with the research variables, followed by a selection process based on relevance and source credibility. Data analysis was conducted using content analysis techniques, which involved stages of data reduction, categorization, synthesis, and inductive conclusion drawing. The findings indicate that the Extreme Learning Machine method demonstrates high accuracy and efficient processing speed in detecting myopia, while the Android platform offers accessibility and ease of use for users. The primary risk factors identified include excessive screen time and limited outdoor activity. Furthermore, eye exercise therapy has been shown to improve visual acuity and reduce eye strain. The integration of these components results in a comprehensive system that functions not only as a detection tool but also as an educational and preventive medium. This study contributes to the field of education by providing a technology-based health education tool and to the scientific community through the advancement of artificial intelligence applications in healthcare. Future research is recommended to extend this study through experimental approaches or direct system implementation to empirically evaluate its effectiveness in real-world settings.