Combining wearable and environmental sensing into an unobtrusive tool for long-term sleep studies

Abstract

Long-term sleep monitoring of patients has been identified as a useful tool to observe sleep trends manifest themselves over weeks or months for use in behavioral studies. In practice, this has been limited to coarse-grained methods such as actigraphy, for which the levels of activity are logged, and which provide some insight but have simultaneously been found to lack accuracy to be used for studying sleeping disorders [8]. This paper presents a method to automatically detect the user's sleep at home on a long-term basis. Inertial, ambient light, and time data tracked from a wrist-worn sensor, and additional night vision footage is used for later expert inspection. An evaluation on over 4400 hours of data from a focus group of test subjects demonstrates a high recall night segment detection, obtaining an average of 94%. Further, a clustering to visualize reoccurring sleep patterns is presented, and a myoclonic twitch detection is introduced, which exhibits a precision of 74%. The results indicate that long-term sleep pattern detections are feasible.

From Wearable Sensors to Smart Implants-–Toward Pervasive and Personalized Healthcare

“…The objective is to reduce the uncertainty of predictions by fusing multimodal information and/or providing a sense of context-awareness, which can improve the level of integration of the application with the monitored scenario. For instance, a sleep disorder monitoring was developed in [11] using a combination of wearable, light sensors and video recordings in order to detect the most relevant events during sleep and allow long-term monitoring. In another example, a fall detection system using information from wearable motion and ambient vision sensor as well as energy consumption (appliances and lights turned on and off) was able to appreciate the context of a fall in order to recognise environmental hazards [12].…”

Capacitive near-field communication for ubiquitous interaction and perception

“…Various approaches led to a wide variety of sensor-based approaches, for example to recognize activities of daily living [1,18,30] or studying various types of sleeping behavior [2,27].…”

A general framework for never-ending learning from time series streams

“…In some situations, P may be a companion to S. For example, in [9], which manually attempts some of the goals of this work, S is a night-vision camera recording sleeping postures and P is a time series stream from a sensor worn on the wrist of the sleeper. In other cases, P could be a transform or low-dimensional projection of S. In one example we consider in our experimental section, S is a stereo audio stream recorded at 44,100Hz, and P is a single-channel 100Hz Mel-frequency cepstral coefficient (MFCC) transformation of it.…”