Sensor-based Sleep Stage Classification Using Deep Learning

Más información sobre el libro

Sleep is a crucial physiological phenomenon essential for human growth and development, akin to diet and exercise. Increasingly, various factors like stress and metabolic disorders contribute to a rise in sleep disorders among individuals. Recent research highlights the importance of sleep stage analysis for early detection and treatment of these disorders. Symptoms vary across age groups, with children experiencing a higher prevalence of sleep disorders compared to adults. While adult sleep stage classification is well-established, children's sleep stages exhibit distinct characteristics, necessitating specialized classification methods for this demographic. The advent of intelligent computing technology has enabled the application of artificial intelligence in medical research, particularly in sleep medicine. Deep learning techniques can efficiently identify relevant sleep features from collected data, allowing for accurate interpretation of children's sleep stages. This approach not only streamlines data analysis by reducing manual annotation efforts but also minimizes the risk of misdiagnosis by sleep experts. This work introduces advanced deep learning methods for classifying sleep stages in children using time series polysomnography data from clinical sensors. The enhanced performance in identifying sleep stages in children with disorders underscores the promising collaboration between artificial intelligence and sleep m