SleepSight: Sleep Disorder Risk Assessment and Recommendations

Abstract

Sleep disorders such as insomnia, sleep apnea, narcolepsy and restless leg syndrome have been on the rise due to reasons like stress, irregular lifestyle habits and lack of public awareness. Early detection is the cure, but most contemporary solutions track only sleep patterns without providing accurate risk estimates, hence creating a vast chasm in preventive healthcare. SleepSight is a web application based on machine learning technology that estimates the likelihood of different sleep disorders based on demographic and lifestyle details like age, gender, sleep duration, stress and activity. The architecture encompasses data preprocessing, optimized multi-class classification and lean architecture to provide accurate and rapid predictions. A set of algorithms like Random Forest, XGBoost, Logistic Regression and Support Vector Machines will be tested and the most efficient model will be implemented. Users are provided with percentage-based risk scores and personalized improvement tips through an easy-to-use interface. The process facilitates proactive lifestyle intervention, encourages frequent self-checks and on-time medical consultations as and when needed. The target users are individuals, wellness professionals, telemedicine practitioners and corporate wellness programs with a larger goal of sensitizing the public, enabling early intervention and encouraging overall sleep wellness at community level