A Structured Learning Approach with Neural Conditional Random Fields for Sleep Staging

Abstract

Sleep plays a vital role in human health, both mental and physical. Sleep disorders like sleep apnea are increasing in prevalence, with the rapid increase in factors like obesity. Sleep apnea is most commonly treated with Continuous Positive Air Pressure (CPAP) therapy. Presently, however, there is no mechanism to monitor a patient's progress with CPAP. Accurate detection of sleep stages from CPAP flow signal is crucial for such a mechanism. We propose, for the first time, an automated sleep staging model based only on the flow signal.Deep neural networks have recently shown high accuracy on sleep staging by eliminating handcrafted features. However, these methods focus exclusively on extracting informative features from the input signal, without paying much attention to the dynamics of sleep stages in the output sequence. We propose an end-toend framework that uses a combination of deep convolution and recurrent neural networks to extract high-level features from raw flow signal with a structured output layer based on a conditional random field to model the temporal transition structure of the sleep stages. We improve upon the previous methods by 10% using our model, that can be augmented to the previous sleep staging deep learning methods. We also show that our method can be used to accurately track sleep metrics like sleep efficiency calculated from sleep stages that can be deployed for monitoring the response of CPAP therapy on sleep apnea patients. Apart from the technical contributions, we expect this study to motivate new research questions in sleep science.

The future of sleep health: a data-driven revolution in sleep science and medicine

“…Data on sleep can be also obtained from devices to treat sleep apnoea, such as Continuous Positive Air Pressure (CPAP). For example, Aggarwal et al 48 showed that CPAP can be used to classify and track sleep metrics, which could be used to monitor the response of CPAP therapy in sleep apnoea patients.…”

Classification of EEG signals using machine learning and deep learning techniques

“…If x={x1, x2, x3, …..xn} are features in RF domain, y be the class label, encoding vector and is given by {y (1) ,y (2) , y (3) , ….y © } where c is the number of class labels. The linear discriminate function of each class is as in Equation (1).…”

“…Fuzzy neural network may be used to identify K-complex in sleep EEG and diagnose neurological, mental abnormalities in [2]., K-complex is a transitory wave with non-linear and dynamic properties. Convolutional neural network (CNN) and deep learning algorithms are utilized on neurological illnesses to level or degree of depression in patients by [3]. , Right hemisphere and left hemisphere are evaluated and hyper activeness of right hemisphere is the degrading component, it is analyzed and depression severity index (DSI) is created.…”

Computational Sleep Behavior Analysis: A Survey

“…Because it is difficult to distinguish N1 and N2, the stages are occasionally fused to light sleep and compared to deep sleep (N3) [96]. This results in 4 stages: wake, REM, light, and deep sleep.…”

“…This results in 4 stages: wake, REM, light, and deep sleep. In [96], continuous positive air pressure (CPAP) flow signals from 400 subjects were analyzed to detect sleep stages. High-level features were extracted with CNN and RNN, which were further used in a conditional random field (CRF).…”