Marko Borazio

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

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. This paper presents a method to automatically detect the users 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 re-call 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.

Already up? using mobile phones to track & share sleep behavior

Users share a lot of personal information with friends, family members, and colleagues via social networks. Surprisingly, some users choose to share their sleeping patterns, perhaps both for awareness as well as a sense of connection to others. Indeed, sharing basic sleep data, whether a person has gone to bed or waking up, informs others about not just ones sleeping routines but also indicates physical state, and reflects a sense of wellness. We present Somnometer, a social alarm clock for mobile phones that helps users to capture and share their sleep patterns. While the sleep rating is obtained from explicit user input, the sleep duration is estimated based on monitoring a users interactions with the app. Observing that many individuals currently utilize their mobile phone as an alarm clock revealed behavioral patterns that we were able to leverage when designing the app. We assess whether it is possible to reliably monitor ones sleep duration using such apps. We further investigate whether providing users with the ability to track their sleep behavior over a long time period can empower them to engage in healthier sleep habits. We hypothesize that sharing sleep information with social networks impacts awareness and connectedness among friends. The result from a controlled study reveals that it is feasible to monitor a users sleep duration based just on her interactions with an alarm clock app on the mobile phone. The results from both an in-the-wild study and a controlled experiment suggest that providing a way for users to track their sleep behaviors increased user awareness of sleep patterns and induced healthier habits. However, we also found that, given the current broadcast nature of existing social networks, users were concerned with sharing their sleep patterns indiscriminately.

Sustained logging and discrimination of sleep postures with low-level, wrist-worn sensors

We present a study which evaluates the use of simple low-power sensors for a long-term, coarse-grained detection of sleep postures. In contrast to the information-rich but complex recording methods used in sleep studies, we follow a paradigm closer to that of actigraphy by using a wrist-worn device that continuously logs and processes data from the user. Experiments show that it is feasible to detect nightly sleep periods with a combination of light and simple motion and posture sensors, and to detect within these segments what basic sleeping postures the user assumes. These findings can be of value in several domains, such as monitoring of sleep apnea disorders, and support the feasibility of a continuous home-monitoring of sleeping trends where users wear the sensor device uninterruptedly for weeks to months in a row.

Enhancing Accelerometer-Based Activity Recognition with Capacitive Proximity Sensing

Activity recognition with a wearable accelerometer is a common investigated research topic and enables the detection of basic activities like sitting, walking or standing. Recent work in this area adds different sensing modalities to the inertial data to collect more information of the user’s environment to boost activity recognition for more challenging activities. This work presents a sensor prototype consisting of an accelerometer and a capacitive proximity sensor that senses the user’s activities based on the combined sensor values. We show that our proposed approach of combining both modalities significantly improves the recognition rate for detecting activities of daily living.

Improving activity recognition without sensor data: a comparison study of time use surveys

Wearable sensing systems, through their proximity with their user, can be used to automatically infer the wearers activity to obtain detailed information on availability, behavioural patterns and health. For this purpose, classifiers need to be designed and evaluated with sufficient training data from these sensors and from a representative set of users, which requires starting this procedure from scratch for every new sensing system and set of activities. To alleviate this procedure and optimize classification performance, the use of time use surveys has been suggested: These large databases contain typically several days worth of detailed activity information from a large population of hundreds of thousands of participants. This paper uses a strategy first suggested by [16] that utilizes time use diaries in an activity recognition method. We offer a comparison of the aforementioned North-American data with a large European database, showing that although there are several cultural differences, certain important features are shared between both regions. By cross-validating across the 5160 households in this new data with activity episodes of 13798 individuals, especially distinctive features turn out to be time and participants location. Additionally, we identify for 11 different activities which features are most suited to be used for later on activity recognition.

Towards Benchmarked Sleep Detection with Wrist-Worn Sensing Units

The monitoring of sleep by quantifying sleeping time and quality is pivotal in many preventive health care scenarios. A substantial amount of wearable sensing products have been introduced to the market for just this reason, detecting whether the user is either sleeping or awake. Assessing these devices for their accuracy in estimating sleep is a daunting task, as their hardware design tends to be different and many are closed-source systems that have not been clinically tested. In this paper, we present a challenging benchmark dataset from an open source wrist-worn data logger that contains relatively high-frequent (100Hz) 3D inertial data from 42 sleep lab patients, along with their data from clinical polysomnography. We analyse this dataset with two traditional approaches for detecting sleep and wake states and propose a new algorithm specifically for 3D acceleration data, which operates on a principle of Estimation of Stationary Sleep-segments (ESS). Results show that all three methods generally over-estimate for sleep, with our method performing slightly better (almost 79% overall median accuracy) than the traditional activity count-based methods.

Using time use with mobile sensor data: a road to practical mobile activity recognition?

Having mobile devices that are capable of finding out what activity the user is doing, has been suggested as an attractive way to alleviate interaction with these platforms, and has been identified as a promising instrument in for instance medical monitoring. Although results of preliminary studies are promising, researchers tend to use high sampling rates in order to obtain adequate recognition rates with a variety of sensors. What is not fully examined yet, are ways to integrate into this the information that does not come from sensors, but lies in vast data bases such as time use surveys. We examine using such statistical information combined with mobile acceleration data to determine 11 activities. We show how sensor and time survey information can be merged, and we evaluate our approach on continuous day-and-night activity data from 17 different users over 14 days each, resulting in a data set of 228 days. We conclude with a series of observations, including the types of activities for which the use of statistical data has particular benefits.

Wear is Your Mobile? Investigating Phone Carrying and Use Habits with a Wearable Device

This article explores properties and suitability of mobile and wearable platforms for continuous activity recognition and monitoring. Mobile phones have become generic computing platforms, and even though they might not always be with the user, they are increasingly easy to develop for and have an unmatched variety of on-board sensors. Wearable units in contrast tend to be purpose-built, and require a certain degree of user adaptation, but they are increasingly used to do continuous sensing. We explore the trade-offs for both device types in a study that compares their sensor data and that explicitly examines how often these devices are being worn by the user. To this end, we have recorded a dataset from 51 participants, who were given a wrist-worn sensor and an app to be used on their Smartphone for two weeks continuously, totalling 638 days (or over 15300 hours) of wearable and mobile data. Results confirm findings of previous studies from North America and show that Smartphones are on average being on their user less than 23% of the time, mostly during working hours. Just as noteworthy is the high variance in Smartphone use (in carrying, interacting with, and charging the phone) among participants.

Predicting sleeping behaviors in long-term studies with wrist-worn sensor data

This paper conducts a preliminary study in which sleeping behavior is predicted using long-term activity data from a wearable sensor. For this purpose, two scenarios are scrutinized: The first predicts sleeping behavior using a day-of-the-week model. In a second scenario typical sleep patterns for either working or weekend days are modeled. In a continuous experiment over 141 days (6 months), sleeping behavior is characterized by four main features: the amount of motion detected by the sensor during sleep, the duration of sleep, and the falling asleep and waking up times. Prediction of these values can be used in behavioral sleep analysis and beyond, as a component in healthcare systems.

Characterizing sleeping trends from postures

We present an approach to model sleeping trends, using a light-weight setup to be deployed over longer time-spans and with a minimum of maintenance by the user. Instead of characterizing sleep with traditional signals such as EEG and EMG, we propose to use sensor data that is a lot weaker, but also less invasive and that can be deployed unobtrusively for longer periods. By recording wrist-worn accelerometer data during a 4-week-long study, we explore in this poster how sleeping trends can be characterized over long periods of time by using sleeping postures only.