PRECIOUS Living Lab Demo @Experiment’HAAL

The living lab Experiment’Haal (Human Ambient Assisting Living) is used in PRECIOUS to realize end-users tests.
A living lab is a laboratory where innovations are proposed to users in a controlled environment.
Experiment’Haal offers an ecosystem to the PRECIOUS project in order to test individual component such as environmental sensors, body weight scale, user feedback, etc.

Below a short video to show the results of our first deployment. Enjoy!

Demo Precious project in the Experiment’HAAL living lab from ptanguy on Vimeo.

Social Media Mood Predictor

 

Nowadays many people worldwide use social media like Facebook, Twitter and the like. They post, comment and share information about themselves or topics they are interested in.

In the context of PRECIOUS we investigate: Can this data be used, to tell more about a person’s mood?

For instance in [2] a web-based tool called ‘MoonPhrases’ was created to enable Twitter users to reflect about their mood and well-being. A similar approach was taken in [1], it was investigated to improve the classification of Tweets in either positive, neutral or negative sentiment. Moreover in [3] and [4] messages of Twitter users were interpreted to find out how users talk about depression in Tweets and how the usage of sentiment words of a depressed person differ from a not depressed person.

1

Therefore our primary hypothesis is that it is possible to leverage mood/emotions from social media messages. The first step consists in collecting emotionally classified messages to build a classifier. As the range of possible emotions is very wide, it is important to choose predefined emotion classes, to limit the resources needed for data collection and processing of the classifier.  For example the emotion stress is very important in context of cardiovascular diseases, which are in the focus of the PRECIOUS project. Therefor a big focus is on collecting messages classified as stressed or not stressed. These classified messages are used to train a statistical model, which can determine if the writer of the message was stressed. Furthermore as a general indicator for well-being the emotions happy and sad, which correlate well with the pleasure dimension of the circumplex affect model, are in the focus here for emotion recognition.

References

[1] Hagen, M.; Potthast, M.; Büchner, M. & Stein, B.,Hanbury, A.; Kazai, G.; Rauber, A. & Fuhr, N. (Eds.), Twitter Sentiment Detection via Ensemble Classification Using Averaged Confidence Scores, Advances in Information Retrieval, Springer International Publishing, 2015, 9022, 741-754

[2] de Choudhury, M.; Gamon, M.; Hoff, A. & Roseway, A. Osmani, V.; Campbell, A. T. & Lukowicz, P. (Eds.), “Moon Phrases”: A Social Media Faciliated Tool for Emotional Reflection and Wellness, Pervasive Computing Technologies for Healthcare (PervasiveHealth), 2013 7th Int. Conference on, 2013

[3] de Choudhury, M.; Gamon, M.; Counts, S. & Horvitz, E., Predicting Depression via Social Media, Seventh International AAAI Conference on Weblogs and Social Media, 2013

[4] Minsu Park, Chiyoung Cha, Meeyoung Cha, and Yoorim, Kweon. Depressive moods of users portrayed in twitter. Telecommunications Review, Jun. 2013.

Sharing Experiences in Design of a Health-care Wearable Device

Monitoring and recommendation of healthy lifestyles is often challenging and requires the correlation between different health-related parameters for a particular individual. On the one hand, some of the lifestyle and biometric data inputs are only possible to be determined by 24/7 monitoring, such as physical activity tracking, sleep quality or food intake nutritional information. On the other hand, it is a big data problem as the correlation between all the inputs is rather challenging and time-consuming to be done by a human being.

Wearable devices are increasingly becoming one of the most common options for continuous monitoring and extraction of an individual’s biometric, lifestyle and activity data. Accelerometer sensors integrated in a wearable device can be used to track the daily physical activity or sleep quality while built-in gyroscope sensor can recognize pattern, such as eating, drinking or taking a bite. Furthermore, heart-rate sensors track the physical activity level or detect stressful situations. Moreover, built-in cameras are able to take a photo of a food and send it to the cloud for further processing, detection and recognition.

A wide range of commercial wearable devices now exist in the market and they are expected to be a booming industry over the next few years. There is also in increased trend of wearable platforms created for developer communities and researchers. Within the PRECIOUS project we have also initiated the design and fabrication of an experimental wearable wristband device to be able to integrate sensors which may not be available in commercial offerings and have full control to the data from the sensors in the wearable. In this article we share for the wider community some of the processes in the development of the PRECIOUS project´s wearable device.

It all started by testing the sensors on a prototype board like the one on Figure 1. The data was processed with an Arduino microcontroller board and sent to a smartphone via Bluetooth. Once the operation of the first prototype was verified, the development was taken into the next step: the Printed Circuit Board (PCB) design and fabrication of the wearable.

 

2015-10-28 10.13.14

Figure 1 Wearable device circuit on a prototype board

The electronic circuit layout of the device was designed and fabricated in home environment for verification. The circuit was then redesigned twice, ensuring the proper operation of the final device. Finally, the PCB of the wearable was fabricated in a laboratory environment in the Aalto University Design Factory and all the components were soldered (see Figure 2).

layout_v102a2015-10-29 15.17.22

Figure 2 Wearable sensors on printed circuit board (PCB layout shown on the left)

In order to update the software of the wearable and perform some tests, temporary wires were directly soldered to the heart of the device (the microcontroller) as shown in Figure 3.

 

pcb_final_programming

Figure 3 Microcontroller programmer hooked to the wearable sensor board.

Finally, a custom-made strap was attached to enable the wearable device to be attached to the wrist as shown in Figure. The device operation was tested by sending all the collected data (heart rate, accelerometer etc.) from the sensors to a smartphone via Bluetooth connection. The experience of this wearable design process has shown that such devices can be conveniently produced by researchers and hobbyists using relatively low-cost components (total cost if the components here was around 20 euros), whilst producing useful readings for research and for those with curiosity in this aspect of quantified self.

final_proto_abovefinal_proto_below

 

 

 

 

 

 

Figure 4. Demonstration of the final design of the PRECIOUS wearable prototype.

From data to understanding yourself

In recent years, the world has lived in the “wellness boom”. People have increasingly been interested in self-measurements and in the market a huge amount of products and applications that allow people to try to feel better has become available. Pedometers, activity trackers, heart rate monitors, optical pulse devices, smart watches etc. allow the collection of more and more continuous knowledge of human behavior. Quantified Self activists have been at the forefront of testing new devices and applications, and bringing these noticeable to the general public.

However, diseases associated with unhealthy lifestyles such as obesity, type 2 diabetes and cardiovascular diseases have shown a steep slope upward in western countries. The question is how the potential of today’s technology can be translated into real benefits, and the thoughts of lifestyle changes into concrete actions. Another important factor to acknowledge regarding measurement data is also its reliability to be really useful. Studies have shown that the devices may show very different values, and for a consumer it can be difficult to know how reliable and accurate the readings in different applications are. Still, only on the basis of reliable data, the conclusions may be relevant and right.

To have a tangible change, we must remember the old but still highly relevant cliché about how the human being is by nature a psychophysiological and social creature, if not even more complex. Consequently, we need a comprehensive grip on what exactly I should do today to be good to myself and why. The PRECIOUS will be a system that is built on reliable sensor data. There the physiological, measured data is combined with knowledge of the user’s psychological condition and motivation. Behind everything is the virtual model build with advanced mathematical and signal processing methods from the person’s autonomic nervous system regulation as measured from heart rate variability. The heart, the most important organ to preserve life, can provide us huge amount of knowledge when it is constantly adjusting its function based on the body’s internal and external needs, day in day out for the whole journey of our lives. This knowledge, among other sensor data, is turned into observations of individual lifestyles, threats, and supportive elements of the user’s well-being. The information further affects the operation of the PRECIOUS mobile application so that the right kinds of challenges and tasks can be provided to the person at the right time, all built on proven psychological methods to support lifestyle changes.

As the development of technologies continues its fast pace, it eventually becomes possible for a person to follow his/her own well-being related parameters 24/7/365 rather ubiquitously. This makes it easier to reach also the most important goal for measuring and collecting data – to learn about ourselves. Because in the end that is what wellbeing is about: knowing and understanding ourselves and the small choices and actions we make today and tomorrow.

Human intervention study investigating the usability of mobile health tools to monitor food intake and physical activity for different user groups.

As part of the PRECIOUS project Campden BRI are conducting a study to investigate the usability of mobile health tools to monitor food intake and physical activity for different user groups.

Project timescale: From 01 August, 2015 to 30 January, 2016

Source of funding: European Union’s Seventh Framework Programme for research, technological development and demonstration under grant agreement no 611366.
Aims: The objective of this intervention study is to investigate the usability of two freely available mobile health tools currently on the market, and one mobile health tools currently being developed through a EC-FP7 project. The study will look at the factors that affect the usability of each mobile health tool for different user groups.

Research project description: This intervention study is being conducted as part of the wider project ‘PREventive Care Infrastructure based on Ubiquitous Sensing’ (PRECIOUS), which has received funding from the European Union’s Seventh Framework Programme for research, technological development and demonstration under grant agreement no 611366. Overall, PRECIOUS aims to provide a preventive care system to promote healthy lifestyles, and as part of this PRECIOUS aims to develop a new monitoring tool, and use motivational techniques for educating the user about healthy eating practices. The objective of this intervention study is to investigate the usability of two freely available mobile health tools currently on the market, and one mobile health tools currently being developed through a EC-FP7 project. The study will look at the factors that affect the usability of each mobile health tool for different user groups (Age 18 to 35 years, with no children living at home, Age 18 to 65 years, with at least one child under 18 years living at home, Age 36 to 60 years, with no children living at home, and Age 61 to 75 years, with no children living at home) in order to improve the design of the EU PRECIOUS mobile health tool. In total three hundred respondents will be recruited (100 per mobile health tool) to complete a 7 day user trial of one of the three mobile health tools. They will then be asked to complete an on-line survey about the usability of the mobile health tool. The outcomes from the study will be used to improve the mobile health tool currently being developed by PRECIOUS.
Results of the study will be reported on the PRECIOUS website; http://www.thepreciousproject.eu/
For further information on the study, please contact:

Miss Charlotte Holmes
Chief Investigator
Campden BRI
GL55 6LD
charlotte.holmes[AT]campdenbri.co.uk
01386 842257

Toward a transparent sensors/actuators layer at home

At home, the PRECIOUS system will offer services to end users: monitoring of the home environment factors (e.g., air quality), gathering data from “health & wellbeing” devices (e.g., body weight scale) and sending feedbacks on home devices (e.g., lamp blinking, messages on TV screen or computer screen, etc.).

However, one of the key issues in the home automation domain is the interoperability between heterogeneous devices and protocols. It might be a problem for end users but also for developers of new applications or services like in the PRECIOUS system.

xAAL_bigPicture

Nowadays, there is a lack of standardization in the home automation systems. In the past ten years, a lot of initiatives, alliances or standards appeared like ZigBee, EnOcean or Z-Wave but they still do not propose interoperability with other alliances or standards. In the same time, the Ambient Assisted Living domain has proposed different solutions mainly based on IP to fight interoperability issues between heterogeneous devices. For more details, we refer the reader to the deliverable D4.1 of the PRECIOUS project.

Telecom Bretagne has proposed xAAL in order to allow the communication between different home automation devices (sensors/actuators) in a home area network. xAAL is a distributed infrastructure for heterogeneous ambient devices. The communication between devices is realized by a lightweight xAAL messaging protocol working on the so-called xAAL bus. The whole infrastructure is based on well-known existing protocols: multicast IPv4/IPv6 UDP.

xAAL will offer the opportunity to use together devices from different manufacturers and technologies. xAAL will also provide a transparent sensors/actuators layer for developers of applications, e.g. well-being application with PRECIOUS. The multiplicity of low-level home automation protocols are hidden by an upper-level generic interface.

How to share information between a smart home and the PRECIOUS platform

The interaction between a smart home and the PRECIOUS system is organized in the two directions: the communication from the user’s home to the PRECIOUS server (uplink), and the communication from the PRECIOUS server to the user’s home (downlink). To realize this uplink/downlink communication, a publish/subscribe protocol is mandatory. For this task, MQTT, a M2M communication protocol, appeared to be well adapted.

The uplink allows to push all data considered as useful (by the user) to the PRECIOUS system. The figure below describes the architecture. The MQTT-xAAL relay sends data from the smart home. The MQTT broker allows to dispatch data to subsribers, e.g. the PRECIOUS server. Then, the MQTT-logger stores data in the PRECIOUS database.

xaal-mqtt-uplink

The downlink will be used by PRECIOUS to activate the scenarios inside the user home or to advertise the user with notifications also called user feedbacks. The MQTT-xAAL relay “subscribes” to PRECIOUS home-scenarios (i.e. topics) . The PRECIOUS scenarios are for example : « blink lamp in the living room », « PRECIOUS notification on TV screen », etc. The MQTT-actuator is a component in the PRECIOUS Cloud Server allowing activating PRECIOUS home-scenarios from PRECIOUS applications.

xaal-mqtt-downlinklink

xAAL is a solution open to the community. For more details, please have a look to our project webpage: http://recherche.telecom-bretagne.eu/xaal/.

Public access on our repository:

svn co https://svn.telecom-bretagne.eu/svn-public/xAAL/

Motivated contributors are invited to contact us for a read-write access on:

svn co https://svn.telecom-bretagne.eu/repository/xAAL/

Finally, all developments for the project will be shared on the githbub repository of the PRECIOUS project:

https://github.com/preciousproject

Encouraging healthier food choices and the role of technology

Over the last few decades, the obesity epidemic has continued to rise. Now in the UK a total of 23% of adults and 33.3% of 10–11-year-olds are considered to be obese. In the UK obesity costs the health care system more than £5bn each year.  Excess weight is a major risk factor for non-communicable diseases such as type 2 diabetes, cancer and heart disease, and it is well-known that these lifestyle-related diseases are now a leading cause of death globally. Diet is a key modifiable risk factor of obesity and therefore non-communicable diseases. As a result, tackling the obesity epidemic in high on the agenda for countries, not only to improve the health of their nation but to reduce the cost to their public healthcare systems and improve economic development. A whole host of approaches are being used across the EU and wider, to encourage people to make healthier food choices, whether it is through education at a school level, increasing the availability of healthier choices or using technology to track individual’s intake and encourage them to make healthier choices.

Many studies have been carried out to understand what influences a person’s choice around food and it has been shown that the average person makes up to 200 food related decisions a day. Wansink, B. (2015) reviewed over a 100 studies where consumer’s behaviours around healthy eating were investigated. From these studies, three main factors were determined to influence an individual’s decision; convenience, attractiveness, and the selection of the healthier choice being the norm. As well as these factors it is also important to recognise the impact of;1) education on why certain choices are the healthier choice and 2) the environment in which these choices are made, have. Therefore, as well as using the factors Wansink, B. (2015) identified when encouraging healthier choices it is critical to ensure there is education incorporated into the approach used and to ensure the behaviours learnt become long term rather than short term.

At PRECIOUS we are using technology to aid users to make healthier lifestyle choices. There are many benefits to the use of technology, including, ease of integration already in daily life, the high level of connectivity which allows the easy transfer of data, and the portability allowing it to be use at any time, any place. As a result, it is likely technology will continue to have a significant role in encouraging healthier food choices. It is nevertheless, important to remember the role personalisation also has in encouraging healthier food choices, during the development of these new technologies, particularly in terms of the nutritional requirements and motivational feedback. Therefore any technology developed will not only need to fit into that’s user’s life with minimal disruption, but will be able to carry out the required level of data processing to personalise its interaction with the user.  This level of personalisation can create challenges due to the number of factors influencing the personalisation required, however through the use of state of the art modelling and design, this is achievable.
blogimage

As part of the PRECIOUS project we aim to encourage healthier food choices, and in turn reduce the risk of type 2 diabetes and cardiovascular disease. To achieve this we are developing a dietary intake and physical activity application to provide individuals with an alternative way to recording their dietary intake. This app will then be part of an overall PRECIOUS system which will use state of the art motivational and gaming techniques to encourage users to change their behaviours to make healthier lifestyle choices. To ensure the application developed makes a real impact on its users, we are carrying out a consumer study to investigate what factors influence the usability of an application for different user groups.  During this study we will investigate two freely available applications and compare them to the current version of the PRECIOUS intake app and another dietary intake app being developed as part of another EU FP7-funded project called; QUALIFY. The study is expected to start in late September 2015, with the results being used to evaluate the current usability of the PRECIOUS application with super users, and then aid the further development.

 References

  1. https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/213720/dh_130487.pdf
  2. Wansink, B. (2015), Change Their Choice! Changing Behavior Using the CAN Approach and Activism Research. Psychol. Mark., 32: 486–500. doi: 10.1002/mar.20794

Towards a motivational framework

“If you want to change attitudes, start with a change in behaviour”
William Glasser

There is a growing body of evidence that demonstrates the potential of mobile communications to radically improve healthcare services. A vast number of applications already exist for different health conditions, but the majority offer similar functions and fail to include a comprehensive motivational framework and sufficient psychological parameters to ensure certain engagement and mid-long term adherence to the service.

Type 2 diabetes (T2D) is a complex disease with an increasing prevalence worldwide. Given this complexity, multifactorial intervention is necessary to improve long-term outcomes as stated in treatment guidelines. From a multidisciplinary perspective, psychological interventions have been utilized to favour better adjustment to the disease and all its implications. Among these psychological approaches, motivational interviewing (MI) has recently become a topic of great interest in the diabetes behavioural field, specifically, to address adherence to guidelines and behaviour change.

MI is a collaborative counselling strategy that engages people in “conversations” that guide them toward strengthening their motivation to change behaviours. It is based on the following assumptions:

  • Ambivalence about change is normal and constitutes an important motivational obstacle in improvement or recovery from a certain situation.
  • Ambivalence can be resolved by working with the individual’s intrinsic motivations and values.
  • An empathic, supportive, yet directive, counselling style provides conditions under which change can occur.

Patients struggling with chronic diseases, such as T2D, require adherence to complex daily regimens and very often, they appear to be only weakly motivated by professionals’ suggestions or other significant ones. Similarly, they could appear poorly motivated to the lifestyle and self-management issues they should address and perhaps, they are not sure on how these improvements should be achieved. A healthcare provider might recommend something such as: “Why don’t you try to exercise a little more and to have lower intake of fats?” The patient response may be reject or resistance: “Yes, but…”, ambivalence: “I’m not sure about…” or hopefully, acceptance: “Yes, you are right, I should…” But despite an initial agreement, lower rates of mid-long term adherence to guidelines are usually observed.Untitled

MI could be very useful in helping patients to assess for themselves their own motives for behaviour change and ultimately, it has been described as a better predictor for final behaviour change. This patient-centred approach is a clear shift away from the previous (and sometimes, somewhere, still very present) paternalistic professional position of feeling responsible for “fixing” the patients’ problems. More and more, it has been described how chronic conditions require a great participation, willingness and self-management abilities from patients. Thus, the “spirit of MI” is more close to a patient empowerment approach, which works perfectly for T2D patients.

The field of technology-supported healthcare is growing rapidly and offers new ways of self-management education and support. Interventions that merge MI with interactive technology may be an efficient and innovative way to address some of these issues because they can be disseminated to new settings, populations, and areas that might not otherwise have the capacity for in-person evidence-based care. MI delivered by new technologies (e.g. mobile applications) can address these issues because the content is programmable and automated (personalized, though), which may be particularly important when disseminating MI in diverse populations and in different languages. This approach is also less expensive than one-on-one treatment, offers easy access, and the anonymity overcomes the stigma sometimes associated with formal treatment.

 

There are not too many mobile applications developed with a well-established and implemented motivational framework from the very initial phases of service design. PRECIOUS service is intended to overcome such limitation and is aimed to combine a multidisciplinary scientific corpus of knowledge, nurtured from information and technology communication, engineering, and psychology and mental health sciences. Available evidence-based MI interventions showing positive results will serve as a reference too.

If you want to know how we are doing so far, you can also read our recently released Newsletter.

References

Image taken from http://www.morguefile.com/archive/display/873584

 

A paradigm shift in digital health

One of the biggest challenges of the 21st century will reside within the healthcare domain: around 60% of the world’s adult population is expected to be overweight by 2030 (Kelly et al. 2008). A sedentary lifestyle, including lack of physical activity and poor choice of diet, can lead to serious chronic diseases burdening public spending in this domain. Given the digital revolution which has entered nearly all markets and industries paired with increasing smartphone penetration rates, developers all over the world have responded to this alarming development by creating mobile applications to deliver health interventions. An estimated 100.000 apps have appeared until today in major mobile app stores, creating what is often referred to as the well-being app sector. However, the underlying problems remain: how do we motivate people to initiate the process of changing behaviour on the one hand, and how do we maintain this behaviour over time in way that people internalize it into their daily habits?

Current health apps are often developed without or only little involvement of health professionals, and are hence only weakly backed by scientific evidence or validation (Bujink et al. 2012, Azar et al. 2013, Boulos et al. 2014). Developers most often present island solutions to specific problems, individually collecting, processing and analysing health data. The high fragmentation of this market is thus one of the key problems in itself: the desired lifestyle change, which often entails a number of different problem areas to address, requires users to install a large number of applications. Finding them on the other hand is an enduring process: users have to try and evaluate them by themselves. There is absolutely no tailoring available to users, where users can find apps based on a match of their own requirements, which can be time or geographical limitations for instance, but also their personality.

At PRECIOUS, we are trying to solve a number of these problems with new approaches. In fact, we want to change the process of how users and developers interact with each other, and to change the way how users select applications on their mobile devices. Our approach is based on a simple assumption: if we were able to recommend apps to users that suit their problems, their situation and their personality, we believe the impact of these applications can be drastically increased. Hence, we are currently creating, designing and implementing a holistic eHealth framework, consisting of a highly-modularized system for so-called gamified applications, also including elements from motivational and behaviour change theory. Developers can finally work together for the benefit of their user, but still find opportunities within our ecosystem for monetizing their efforts. We intend a single point of contact with our users: the PRECIOUS app. In contrast to current solutions however, we intend to run third-party apps within our own app, thus we are able to recommend apps to the users that increase the probability of compliance with a specific treatment suggestions. Image for instance the goal of increasing the user’s physical activity level: current apps set specific goals for the user that are harder to achieve over time, offer gamified aspects that target the motivational side of the problem etc. However, within PRECIOUS, we go one step further: how has this application impacted other users that have used it before? What type of users have responded particularly well to it? Does the user actually have time for e.g. going on a 15 minutes run? How is the weather outside, is it actually suited for this intervention suggestion? All these factors need to be considered when suggestion that specific application.

We believe that if we find the right app, for the right person, at the right time, we can achieve a higher impact upon the individual, and have a lasting influence towards healthier behaviour.

References

  • Kelly, T, W Yang, CS Chen, K Reynolds, and J He. 2008. “Global Burden of Obesity in 2005 and Projections to 2030.” International Journal of Obesity 32 (9). Nature Publishing Group: 1431–37.
  • Buijink, Arthur Willem Gerard, Benjamin Jelle Visser, and Louise Marshall. 2012. “Medical Apps for Smartphones: Lack of Evidence Undermines Quality and Safety.” Evidence Based Medicine. BMJ Publishing Group Ltd, ebmed–d2012.
  • Azar, Kristen MJ, Lenard I Lesser, Brian Y Laing, Janna Stephens, Magi S Aurora, Lora E Burke, and Latha P Palaniappan. 2013. “Mobile Applications for Weight Management: Theory-Based Content Analysis.” American Journal of Preventive Medicine 45 (5). Elsevier: 583–89.
  • Boulos, Maged N Kamel, Ann C Brewer, Chante Karimkhani, David B Buller, and Robert P Dellavalle. 2014. “Mobile Medical and Health Apps: State of the Art, Concerns, Regulatory Control and Certification.” Online Journal of Public Health Informatics 5 (3). University of Illinois at Chicago Library: 229.