Abstract
The quality of sleep has a significant impact on health and life. This study adopts the structure of hierarchical classification to develop an automatic sleep stage classification system using ballistocardiogram (BCG) signals. A leave-one-subject-out cross validation (LOSO-CS) procedure is used for testing classification performance. Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM), and Deep Neural Networks DNNs are complementary in their modeling capabilities; while CNNs have the advantage of reducing frequency variations, LSTMs are good at temporal modeling. A transfer learning (TL) technique is used to pre-train our CNN model on posture data and then fine-tune it on the sleep stage data. We used a ballistocardiography (BCG) bed sensor to collect both posture and sleep stage data to provide a non-invasive, in-home monitoring system that tracks changes in health of the subjects over time. Polysomnography (PSG) data from a sleep lab was used as the ground truth for sleep stages, with the emphasis on three sleep stages, specifically, awake, rapid eye movement (REM) and non-REM sleep (NREM). Our results show an accuracy of 95.3%, 84% and 93.1% for awake, REM and NREM respectively on a group of patients from the sleep lab.
Highlights
Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM), and Deep Neural Networks (DNNs) are individually limited in their modeling capabilities, and we believe that time series data classification can be improved by combining these networks in a unified framework
We propose a method for classifying sleep stages based on the CNN, LSTM and DNN with the help of transfer learning
We have developed a deep learning-based hierarchical classification method for automatic sleep stage classification based on BCG data
Summary
Sleep is a critical physiological phenomenon for recovery from mental and physical fatigue. CNNs, LSTMs, and DNNs are individually limited in their modeling capabilities, and we believe that time series data classification can be improved by combining these networks in a unified framework. Many feature extraction approaches in time series face issues related to the nonstationary nature of the signal when the probability distribution does not change over time. We propose a method for classifying sleep stages based on the CNN, LSTM and DNN with the help of transfer learning. We use a transfer learning technique to train our network model with sleep posture data for 56 subjects (source dataset) and use it for sleep stage classification (target data). The CNNs are trained to learn filters that extract time-invariant features from the BCG signals while the LSTMs are trained to encode temporal information such as sleep stage transition rules
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
