Bidirectional Long Short-Term Memory based Student Common Disease Monitoring System

The increasing prevalence of common diseases among students necessitates an efficient and proactive health monitoring system to ensure timely intervention and improved well-being. Traditional disease monitoring methods often struggle to effectively model temporal dependencies in health data, resulting in suboptimal classification results. To overcome these challenges, this research presents a Deep Learning (DL)-based Student Common Disease Monitoring System utilizing BiLSTM (Bidirectional Long Short-Term Memory) for accurate disease prediction. The system leverages student health data from Kaggle, incorporating key physiological and medical parameters to enhance predictive accuracy. By capturing sequential dependencies in health records, BiLSTM effectively models complex temporal patterns, enabling precise identification of disease onset. Experimental results demonstrate the proposed system’s superior performance, achieving a high classification accuracy of 98.7%, significantly outperforming traditional methods. Key evaluation metrics, including Precision (97.3%), Recall (97.9%), Specificity (98.4%), and F1-Score (97.5%), validate the model's reliability and robustness. These results highlight the practical applicability of the proposed approach in real-world educational settings. These outcomes highlight the ability of AI-driven health monitoring systems in educational settings, supporting early disease detection and proactive healthcare management, thereby contributing to improved public health outcomes.