Early Detection of Child Eye Dryness Using IoT and Deep Learning Enabled AI Model

The increasing usage of mobile phones among chil-dren, especially those born after 2016, has led to a rise in eye dryness disorders. Prolonged exposure to digital screens reduces blink rates, causing inadequate tear film production and increasing the risk of dry eye syndrome. Children born in the digital age (2019 and later) are particularly vulnerable due to early and excessive screen engagement. The lack of awareness among parents and educators about the long-term impact of digital eye strain further exacerbates the problem, leading to potential vision-related disorders in the future. Studies indicate that children who use screens continuously for extended periods experience faster tear evaporation, increasing dryness severity. To address this concern, we propose an IoT-enabled Eye Dryness Detection Framework utilizing deep learning, specifically Long Short- Term Memory (LSTM) models, for accurate classification. The system integrates three core modules: image preprocessing and blink detection, environmental sensor analysis, and AI-based classification. The LSTM model is trained on video images and environmental parameters to classify eye dryness severity into four categories: Normal, Mild, Moderate, and Severe. The model analyzes sequential blink rate variations and tear film breakup time to improve precision. Experimental results validate the efficiency of the proposed system, achieving an impressive 94% accuracy in detecting eye dryness levels. This IoT framework, combined with deep learning, provides real-time monitoring and early intervention, reducing the risk of severe eye disorders among children. This work highlights the urgent need for proactive eye care solutions in the digital era.