Sleep Scan: Smart ML Models For Early Sleep Disorder Identification In Real Time

Abstract

Sleep disorders such as insomnia, sleep apnea, and restless leg syndrome significantly impact an individual’s physical health, cognitive performance, and overall quality of life. However, early diagnosis remains a represents a formidable bottleneck stemming from the sparse distribution of sleep laboratories, high costs, and lack of continuous monitoring solutions. In an effort to reconcile these systemic deficiencies, our proposed work proposes Sleep Scan Smart, a real time machine learning-based analytical platform tailored to the incipient recognition of sleep disorders in a non invasive and cost-effective manner. The system continuously collects physiological signals including heart rate, oxygen saturation (SpO₂), respiration rate, body movement, and electroencephalogram (EEG) data using wearable or sensor-based devices. These signals undergo real-time preprocessing and feature extraction to remove noise and enhance data quality. Advanced machine learning and deep learning models such as Random Forest, Support Vector Machine (SVM), Long Short-Term Memory (LSTM), and Convolutional Neural Networks (CNN) are employed to analyze patterns and for the algorithmic identification of aberrant patterns secondary to distinct sleep disorders. Sleep Scan Smart performs real-time anomaly detection and generates instant alerts, enabling timely intervention. The system also provides comprehensive sleep health insights through a user-friendly interface, helping users and healthcare professionals make informed decisions. By realizing exceptional predictive exactness under highly constrained resource false alarms, the proposed solution ensures reliable monitoring. Overall, the system enables continuous, home-based sleep assessment and contributes to the advancement of smart, AI-driven preventive healthcare by facilitating early diagnosis and improving long-term health outcomes.