Mamba-based Deep Learning Approach for Sleep Staging on a Wireless Multimodal Wearable System Without Electroencephalography.

Key Points

• This research aims to develop a deep learning model to accurately identify sleep stages using a wearable device without EEG.
• Developed a Mamba-based deep learning model.
• Applied model to data from a wireless multimodal wearable system.
• Tested effectiveness on data from adults in a sleep clinic.
• Successfully inferred major sleep stages from the wearable system.
• Eliminated the need for electroencephalography in sleep monitoring.