Background and objective: Sleep disorders, such as insomnia can seriously affect an individual's performance and even lead to psychological problems. Diagnosis and treatment of sleep disorders require the collection of subjective and objective measures of sleep structure. Subjective measures are currently collected during face-to-face clinical consultations with a medical practitioner. Objective measures, obtained from polysomnographic (PSG) electrodes on the patient's body, are used to quantify sleep quality, specifically slow wave sleep (SWS) periods. In this work, we present an unobtrusive and ambulatory insomnia monitoring system which uses a single electro-oculograph (EOG) channel instead of a full PSG to identify SWS periods together with a mobile health application to collect subjective measures from patients. Method: Fig. 1 presents the architecture of our proposed system. It consists of three main components. The EOG sensing device records one EOG channel and wireless transmits it to the coordinator. The Coordinator (mobile phone) collects data from the EOG device and transmits it back to the clinical back-end. It also includes a mobile sleep diary application implemented on the Android OS to collect the subjective sleep assessment data from patients. The recorded data is time and date stamped and can be used for historical data analysis. The Clinical back-end consists of a database to store the received data and a server with a SWS classifier to assess sleep quality. Our SWS classifier is based on a simple rule based algorithm using spectral features extracted from several bands (alpha band (8 - 12 Hz), beta band (18 - 30 Hz) and delta band (0.5 - 4 Hz) with adaptive thresholds. We used EOG overnight sleep recordings from nine healthy subjects (18-64 years) to validate our SWS detector. Results: Our developed sleep diary was compared to existing sleep diaries applications. It was found to allow efficient handling of data with an improved layout and interface that enhanced user experience. Information collected through the diary also provided the clinician with better access to the required subjective data for an accurate diagnosis of insomnia. Our developed SWS detection algorithm was run on a testing group with 5 subjects and a validation group of 4 subjects. Sensitivity, specificity, relative observed agreement and Cohen's kappa coefficient values were computed and used to compare the output of the algorithm to the sleep experts' analysis. The agreement of our SWS detection method for the validation data was 90.0%, sensitivity 90.5%, specificity 89.9% and kappa value 0.74. Conclusions: Current PSG systems with multiple electrodes are inconvenient and uncomfortable for the user, resulting in modified sleep activity different from their normal night of sleep. Our work has shown that it is possible to reduce the complexity of the insomnia monitoring experience and still collect the required subjective and objective information needed to assist in insomnia diagnosis. Acknowledgement: The work was supported by NPRP grant #[5-1327-2-568] from the Qatar National Research Fund which is a member of Qatar Foundation. The statements made herein are solely the responsibility of the authors.


Article metrics loading...

Loading full text...

Full text loading...

This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error