Improving Sleep Quality Analysis in IoT-Based Smart Healthcare Systems with Hunger Games Search Optimization and Deep Learning
P Vedasundara Vinayagam, S Kanagamalliga, Syed Abudahir A, Mohammed Asan Sithik · 2025
Conventional approaches for sleep quality monitoring in IoT-based smart healthcare systems mostly rely on human feature extraction and classical machine learning algorithms, which frequently lead to reduced classification accuracy and inefficiencies in real-time analysis. High processing loads and individual sleep variability are challenges for these systems. The proposed method combines the Hunger Games Search Optimization (HGSO) algorithm with deep learning models to solve these problems and improve sleep analysis accuracy and efficiency. By optimizing hyperparameters, the HGSO algorithm enhances feature extraction and sleep stage categorization. The proposed method receives a higher rating of 9.1 from users, indicating its efficacy in providing proactive sleep management and individualized feedback, even though it performs less accurately in categorization and takes longer to process than traditional methods. The system's cloud-based deployment and real-time data analysis further provide scalability and quick reactions, greatly improving the user experience and healthcare results overall.