context-aware based quality of life telemonitoring

Metadata of the chapter that will be visualized inSpringerLink

Book Title Distributed Systems and Applications of Information Filtering and RetrievalSeries Title 7092

Chapter Title Context-Aware Based Quality of Life Telemonitoring

Copyright Year 2014

Copyright HolderName Springer-Verlag Berlin Heidelberg

Corresponding Author Family Name VargiuParticle

Given Name EloisaSuffix

Division

Organization Barcelona Digital Technology Center

Address Barcelona, Spain

Email [email protected]

Author Family Name FernándezParticle

Given Name Juan ManuelSuffix

Division




Author Family Name MirallesParticle

Given Name FelipSuffix

Division




Abstract Telemonitoring Quality of Life of individuals is the base for current and future telemedicine andteleassistance solutions which will become paramount in the sustainability and effectiveness of healthcaresystems. In the framework of the BackHome European R&D project, which aims to provide atelemonitoring and home support system using Brain Computer Interfaces and other assistive technologiesto improve autonomy and quality of life of disabled people, we propose a methodology to assess andtelemonitor quality of life of individuals based on the awareness of user context. This methodology holds ageneric approach to be applied to other eHealth use cases and is based on the acquisition, fusion andprocessing of heterogeneous data coming from sensors, devices, and user interaction, and the knowledgeinferred from the correlation of this processed data and the input coming from our proposed questionnairesmapped to standard taxonomies. The proposed methodology is very ambitious and although we arepresenting preliminary validation, it will have to be formally validated and enhanced with the study ofrepresentative user data which will be acquired within BackHome and other related projects.

UN

CO

RR

EC

TE

D P

RO

OF

Chapter 1Context-Aware Based Quality of LifeTelemonitoring

Eloisa Vargiu, Juan Manuel Fernández and Felip Miralles

Abstract Telemonitoring Quality of Life of individuals is the base for current and1

future telemedicine and teleassistance solutions which will become paramount in2

the sustainability and effectiveness of healthcare systems. In the framework of the3

BackHome European R&D project, which aims to provide a telemonitoring and home4

support system using Brain Computer Interfaces and other assistive technologies to5

improve autonomy and quality of life of disabled people, we propose a methodology6

to assess and telemonitor quality of life of individuals based on the awareness of user7

context. This methodology holds a generic approach to be applied to other eHealth8

use cases and is based on the acquisition, fusion and processing of heterogeneous9

data coming from sensors, devices, and user interaction, and the knowledge inferred10

from the correlation of this processed data and the input coming from our proposed11

questionnaires mapped to standard taxonomies. The proposed methodology is very12

ambitious and although we are presenting preliminary validation, it will have to be13

formally validated and enhanced with the study of representative user data which14

will be acquired within BackHome and other related projects.15

1 Introduction16

The demographic trend of our ageing society is partly due to the amazing progress17

of medicine in the last decades, which has increased life expectancy and improved18

quality of life, specially to people living in developed countries.19 AQ1

E. Vargiu (B) · J. Manuel Fernández · F. MirallesBarcelona Digital Technology Center, Barcelona, Spaine-mail: [email protected]

J. Manuel Fernándeze-mail: [email protected]

F. Mirallese-mail: [email protected]

C. Lai et al. (eds.), Distributed Systems and Applications of Information Filtering, 1and Retrieval, Studies in Computational Intelligence 515,DOI: 10.1007/978-3-642-40621-8_1, © Springer-Verlag Berlin Heidelberg 2014

310893_1_En_1_Chapter � TYPESET DISK LE � CP Disp.:21/8/2013 Pages: 23 Layout: T1-Standard

Au

tho

r P

roo

f

UN

CO

RR

EC

TE

D P

RO

OF

2 E. Vargiu et al.

However, this demographic shift comes along with an important stress to our20

healthcare systems, which nowadays face sustainability problems.21AQ2

Design of solutions which take advantage of new Information and Communica-22

tion Technologies (ICT) provide efficiency, efficacy and cost-effectiveness to care23

practice. Telemedicine solutions allow treating chronic patients living at home, pre-24

venting and predicting exacerbations and decrease costly hospitalizations. Telereha-25

bilitation solutions enable following continuous interventions which may improve26

health conditions without the need for the patient to physically move to specialized27

facilities. Teleassistance solutions facilitate improving autonomy, safety and social28

participation of people with special needs, namely the elderly and in particular the29

disabled, through home support technologies which postpone socio-sanitary services30

and associated costs.31AQ3

One key common feature of all those novel eHealth solutions is telemonitoring,32

which makes possible to remotely assess health status and quality of life of individu-33

als. By acquiring heterogeneous data coming from sensors (physiological, biometric,34

environmental; non invasive, adaptive and transparent to user) and data coming from35

other sources (e.g., interaction of the user with digital services) to become aware36

of user context; by inferring user behaviour and detecting anomalies from this data;37

and by providing elaborated and smart knowledge to clinicians, therapists, carers,38

families, and the patients themselves, we will be able to foster preventive, predictive39

and personalized care actions, decisions and support.40AQ4

In the context of an assistive environment that provides home support to people41

with disabilities, in this chapter we propose a generic methodology to telemonitor42

quality of life of individuals with a holistic bio-psycho-social approach, which intends43

to become the base for current and future telemedicine and teleassistance solutions.44

The chapter is organized as follows, Sect. 2 resumes main related work concerning45

quality of life assessment, context-aware user profiling, and telemonitoring and home46

support. In Sect. 3, we summarize the objective of the project in which the proposed47

methodology is studied. Section 4 presents and discusses the proposed methodology48

together with preliminary experimental results. Section 5 ends the chapter with some49

conclusions.50

2 Background51

In this chapter, we are interested in presenting a general methodology to telemonitor52

quality of life through a context-aware solution. To give a view of all the related53

issues, in this section, we focus on relevant work on quality of life assessment,54

context-aware user profiling, and telemonitoring and home support.55


Au

tho

r P

roo

f

UN

CO

RR

EC

TE

D P

RO

OF

1 Context-Aware Based Quality of Life Telemonitoring 3

2.1 Quality of Life Assessment56

2.1.1 Definitions57

QoL (sometimes refereed to as Health-Related QoL or HRQoL) is defined by the58

subjective experiences or preferences expressed by an individual, or members of a59

particular group of persons, in relation to specified aspects of health status that are60

meaningful, in definable ways, for that individual or group [50]. According to [19],61

QoL is a state of well-being defined by two components: (i) the ability to perform62

everyday activities, which reflects physical, psychological and social well-being,63

and patient satisfaction with levels of functioning, and (ii) the control of disease64

and treatment symptoms. Also, as Lerer [26] suggests, e-health consumers are now65

empowered by an increased ability to obtain health information via the Internet, with66

the main objective to maintain the highest possible level of QoL.67

The World Health Organization (WHO) defines QoL as the individuals’ percep-68

tion on their position in life within the cultural context and the value system in69

which the individuals live and with respect to their goals, expectations, norms and70

worries [58]. It is a multidimensional and complex concept that includes personal71

aspects, like health, autonomy, independence, satisfaction with life and environmen-72

tal aspects such as support networks and social services, among others. The World73

Health Organization Quality of Life (WHOQOL) project [48] has the aim to develop74

an international, cross-cultural QoL-assessment instrument based on this definition.75

The WHOQOL instrument was collaboratively developed in a number of centers76

worldwide, and has been widely field-tested.77

Patrick et al. [41] define QoL as the value assigned to life duration based on78

the perception of physical, psychological, and social limitations. According to their79

view, QoL is related to the reduction in opportunities due to diseases, their sequel,80

treatment, and to health policies. Naughton et al. [37] define QoL as the subjective81

perception, influenced by the current health status, of the ability to realize activities82

important for the person.83

QoL could also be considered as a dynamic and changing concept that includes84

continuous interactions between the person and the environment. Accordingly, QoL85

in ill people is related to the interaction among the disease, the patients’ character,86

the change in their life, the received social support, as well as the period of life in87

which the disease appears.88

Healthcare organizations use several tools to acquire QoL-related information.89

These tools make use of specific terms, which are sometimes ambiguous: descriptor,90

grade, item, index, indicator, parameter, questionnaire, scale, score, and test. The91

terminology used in this chapter is part of an ontology (and encoded in OWL 2 [22])92

and is defined as follows [9]:93

• Indicator: a (subjective or objective) parameter, category, or descriptor used to94

measure or compare activities and participation, body functions, body structures,95

environment factors, processes, and results (e.g., dressing).96


Au

tho

r P

roo

f

UN

CO

RR

EC

TE

D P

RO

OF

4 E. Vargiu et al.

• Index: a combination of indicators, questionnaires and possibly other indexes. The97

function representing this combination gives as summarizing result a score (e.g.,98

Barthel index).99

• Item: a single question or concept (e.g., Mobility).100

• Questionnaire (or instrument or test): a set of questions (or items) answered using101

a scale (e.g., EQ-5D).102

• Scale: a mapping between some ordered (qualitative or quantitative) values (or103

grades) and their description. These values are used to answer questionnaires (e.g.,104

I have no problems in walking about, I have some problems in walking about, I105

am confined to bed).106

2.1.2 Questionnaires for Assessment of Quality of Life107

Several questionnaires have been proposed and adopted to assess QoL. Let us sum-108

marize here the most widely adopted:109

• The WHOQOL-BREF questionnaire [35] comprises 26 items, which measure the110

following broad domains: physical health, psychological health, social relation-111

ships, and environment.112

• The EQ-5D-5L questionnaire [51] was developed by the EuroQol Group in order113

to provide a generic measure of health status. Applicable to a wide range of health114

conditions and treatments, it provides a simple descriptive profile and a single115

value for health status that can be used in the clinical and economic evaluation of116

healthcare as well as in population health surveys.117

• The RAND-36 questionnaire [21] is comprised of 36 items that assess eight health118

concepts: physical functioning, role limitations caused by physical health prob-119

lems, role limitations caused by emotional problems, social functioning, emotional120

well-being, energy/fatigue, pain, and general health perceptions.121

• The Short Form (36) Health Survey (SF-36v2) [57] is a questionnaire about patient122

health status and is commonly used in health economics in the quality-adjusted123

life year calculation to determine the cost-effectiveness of a health treatment. The124

SF-36 and RAND-36 include the same set of items, however the scoring of general125

health and pain is different [44].126

• The Barthel questionnaire [40] is used to measure performance in Activities of127

Daily Living (ADLs). It uses ten variables describing ADLs and mobility. The128

higher the score derived from this questionnaire, the greater the likelihood of129

being able to live at home with independence following discharge from hospital.130

2.1.3 Existing Standardization Efforts131

Several standard terminologies and classifications exist, which can be used for an132

interoperable representation of QoL. Some examples are: the Systematized Nomen-133

clature of Medicine Clinical Terms (SNOMED CT); the Unified Medical Language134


Au

tho

r P

roo

f

UN

CO

RR

EC

TE

D P

RO

OF


System (UMLS); the International Classification of Diseases version 10 (ICD-135

10); and the International Classification of Functioning, disability and health (ICF)136

defined by the WHO. In addition to terminologies and classifications, information137

models such as the virtual Medical Record (vMR) contribute to solve interoperability138

problems in the electronic exchange of QoL information.139

Several questionnaires are used to evaluate functioning, disability and health. The140

ICF classifies these concepts, specifies their range of values, and can be used to solve141

interoperability problems among health institutions that employ different measuring142

questionnaires. To this aim, questionnaire items can be encoded to ICF concepts143

following the standardization methodology proposed by [10]. Difficulties in mapping144

clinical questionnaires to standard terminologies and ontologies in the rehabilitation145

domain (e.g., data from questionnaires having a finer granularity than ICF categories)146

have been addressed in [9] and [49]. ICF core sets are subsets of the ICF that have147

been created according to specific pathologies or rehabilitation processes. Core sets148

are useful because, in daily practice, clinicians and other professionals can use only149

a fraction of the about 1400 categories found in the ICF.150

2.2 Context-Awareness151

From the first time that the term context-aware computing was introduce by Schilit152

et al. in 1994 [47] several definitions of context have been proposed. Among others,153

let us consider the definition by Dey [14]: “Context is any information that can be154

used to characterize the situation of an entity. An entity is a person, place, or object155

that is considered relevant to the interaction between a user and an application,156

including the user and applications themselves”.157

This definition simplifies the concept of the information related to an interaction;158

avoiding the inclusion of information about other elements that can be at the scenario,159

without influencing the interaction between the user and the application. Following160

this definition, any information related to the involved elements can be used to char-161

acterize the context. In so doing, the context is the conjunction of specific data only162

related to the entities involved in the interaction.163

In order to complement this definition, we follow the classification proposed by164

Zimmerman et al. [59]. This classification takes into account five different cate-165

gories the information of the context, namely “five fundamental categories for con-166

text information”: “Individuality Context”, “Location Context”, “Activity Context”,167

“Relations context” and “Time Context”.168

The “Individuality Context” describes the state of the entity itself, offering the169

information that can be observed about it. This category divides the entities in four170

different types:171

• Natural Entity Context: it contains the entities which appear without the human172

intervention. Include living and inert entities (e.g., atmosphere, water and plants).173

• Human Entity Context: it refers to all the characteristics of human beings (e.g.,174

user’s preferences about language, input device and colour schemes).175


Au

tho

r P

roo

f

UN

CO

RR

EC

TE

D P

RO

OF

6 E. Vargiu et al.

• Artificial Entity Context: as opposite of the Natural Entity Context, it includes all176

the elements developed or built by humans (e.g., buildings, ambient sensors and177

smart home devices).178

• Group Entity Context: it is a collection of entities that share common characteristics179

or have some relation (e.g., a group of people with the same disorder, as “Aquired180

Brain Injury”, or a group of relatives, as “My Family”).181

The “Location Context” includes the information related to the position of the182

entity. It involves the global or relative position among entities, independently from183

the technique used to positioning them. For instance, user’s home can be used as184

spatial information or a coordinated system can be used. Moreover, this information185

can be related to a non-physical position like IP address which is a position a smart186

home device connected to computer network.187

“Activity Context” covers the activities where the entity is, was and will be,188

involved and can be described, for instance, as tasks, aims, and actions.189

The “Relations Context” describes the relations among different entities of a190

context-aware system, such as human beings or things. This information can be191

classified into three kinds of relations: “Social Relations”, “Functional Relations”,192

and “Compositional Relations”.193

Finally, the last category is “Time Context”. In fact, usually the features of the194

context can be evaluated or have variations from one temporal point to other, it means195

that they have a temporal dimension which should be considered as a key information196

for the context [20].197

Other definitions and approaches have been proposed in the literature. In the198

application proposed by Bhattacharyya [4], the following categories have been con-199

sidered:200

• “User information”, which contains knowledge on habits, emotional state, and201

physiological conditions. This category matches with the “Human Entity Context”202

proposed in [59].203

• “Users activities”, which includes spontaneous activity, engaged tasks, or idle204

state. It is quite similar to the “Activity Context”.205

• “Location”, which includes global and relative position, directly matches with206

“Location Context”.207

• “Physical conditions”, which contains light, pressure, heart rate, and temperature.208

It also corresponds to the “Human Entity Context”.209

Summarizing, this approach does not take into account the “Relation Context” infor-210

mation and does not explicitly include the “Time Context”.211

In [6], authors stress the difference between context model and user model by the212

way of obtaining the data. On the one hand, they state that the user model is related to213

data acquired thought the interactions of the user with the application. On the other214

side, the context model is obtained mainly from sensors. Especially this last issue is215

in contradiction from our view. In fact, we consider the context as a complete set of216

information that can come from sensors as well as from interactions and/or relations217

with other entities involved in the same context-aware system.218


Au

tho

r P

roo

f

UN

CO

RR

EC

TE

D P

RO

OF


2.3 Telemonitoring and Home Support219

A simple definition of telemonitoring is the one provided by the Institute of Medicine220

in U.S. [27]: “the remote monitoring of patients including the use of audio, video,221

and other telecommunications and electronic information processing technologies222

to monitor patient status at a distance”.223

As reported in [30], telemonitoring systems have been successful adopted in car-224

diovascular, hematologic, respiratory, neurologic, metabolic, and urologic activities225

[28]. In fact, some of the more common things that telemonitoring devices keep226

track of include blood pressure, heart rate, weight, blood glucose, and hemoglobin.227

Telemonitoring is capable of providing information about any vital signs, as long as228

the patient has the necessary monitoring equipment at her/his location. In principle, a229

patient could have several monitoring devices at home. Clinical-care patients’ phys-230

iologic data can be accessed remotely through the Internet and handled computers231

[46]. Depending on the severity of the patient’s condition, the health care provider232

may check these statistics on a daily or weekly basis to determine the best course of233

treatment.234

In addition to objective technological monitoring, most telemonitoring systems235

include subjective questioning regarding the patient’s health and comfort [28]. This236

questioning can take place automatically over the phone, or telemonitoring software237

can help keep the patient in touch with the health care provider. The health care238

provider can then make decisions about the patient’s treatment based on a combina-239

tion of subjective and objective information similar to what would be revealed during240

an on-site appointment.241

Home sensor technology may create a new opportunity to reduce costs by helping242

people stay healthy and in their homes longer as they age. An interest has therefore243

emerged in using home sensors for health promotion [24]. One way to do this is244

by Telemonitoring and Home Support Systems (TMHSSs). TMHSSs are aimed at245

remotely monitoring patients who are not located in the same place of the health246

care provider. Those supports allow patients to be maintained in their home [12].247

Better follow-up of patients is a convenient way for patients to avoid travel and to248

perform some of the more basic work of healthcare for themselves, thus reducing249

the corresponding overall costs [2, 56].250

Summarizing, a TMHSS allows:251

• To improve the quality of clinical services, by facilitating the access to them,252

helping to break geographical barriers.253

• To keep the objective in the assistance centred in the patient, facilitating the com-254

munication between different clinical levels.255

• To extend the therapeutic processes beyond the hospital, like patient’s home.256

• A saving for unnecessary costs and a better costs/benefits ratio.257

In the literature, several TMHSSs have been proposed. Among others, let us258

recall here the works proposed in [7, 11], and [31]. The system proposed in [7]259

provides users personalized health care services through ambient intelligence. That260


Au

tho

r P

roo

f

UN

CO

RR

EC

TE

D P

RO

OF

8 E. Vargiu et al.

system is responsible of collecting relevant information about the environment. An261

enhancement of the monitoring capabilities is achieved by adding portable mea-262

surement devices worn by the user thus vital data is also collected out of the house.263

Corchado et al. [11] propose a TMHSS aimed at improving healthcare and assistance264

to dependent people at their homes. That system is based on a SOA model for integrat-265

ing heterogeneous wearable sensor networks into ambient intelligence systems. The266

adopted model provides a flexible distribution of resources and facilitates the inclu-267

sion of new functionalities in highly dynamic environments. Sensor networks provide268

an infrastructure capable of supporting the distributed communication needed in the269

dependency scenario, increasing mobility, flexibility, and efficiency, since resources270

can be accessed regardless their physical location. Biomedical sensors allow the sys-271

tem to acquire continuously data about the vital signs of the patient. Mitchell et al.272

[31] propose ContextProvider, a framework that offers a unified, query-able inter-273

face to contextual data on the device. In particular, it offers interactive user feedback,274

self-adaptive sensor polling, and minimal reliance on third-party infrastructure. It275

also allows for rapid development of new context and bio-aware applications.276

As for BNCI users, some work has been presented to provide smart home control277

[15, 17, 23, 52]. To our best knowledge, telemonitoring has not been integrated yet278

with BNCI systems apart as a way to allow remote communication between therapists279

and users [32].280

3 The BackHome Project281

BackHome1 is an EU project concerning physical and social autonomy of people282

with disabilities, by using mainly Brain/Neural Computer Interface (BNCI) and inte-283

grating other assistive technologies as well.284

“BNCI” includes two types of technologies: EEG based Brain Computer Interface285

(BCI) for command and control and affective computing based on EEG activity and286

other physiological signals. BCIs are devices that allow for communication and287

control via thought alone [29, 39, 42]. The term “BNCI” is broader than BCI, since288

BNCIs include systems that sense indirect measures of brain activity, and may not289

provide real-time feedback [5, 33, 43].290

BackHome is partly based on the outcomes coming from BrainAble,2 an EU291

project aimed at offering an ICT-based human-computer-interaction composed of292

BNCI system combined with affective computing, virtual environments and the pos-293

sibility to control heterogeneous devices like smart home environments and social294

networks [38]. BackHome advances BrainAble in supporting the transition from295

institutional care to home post rehabilitation and discharge [13].296

BackHome aims to study the transition from the hospital to the home, focusing on297

how people use BNCIs in both settings. Moreover, it is aimed to learn how different298

1 www.Backhome-FP7.eu2 www.brainable.org


Au

tho

r P

roo

f

www.Backhome-FP7.eu

www.brainable.org

UN

CO

RR

EC

TE

D P

RO

OF


BNCIs and other assistive technologies work together and can help clinicians, dis-299

abled people, and their families in the transition from the hospital to the home. The300

final goal of BackHome is to reduce the cost and hassle of the transition from the301

hospital to the home by developing improved products. To produce applied results,302

BackHome will provide: new and better integrated practical electrodes; friendlier303

and more flexible BNCI software; and better telemonitoring and home support tools.304

Among the overall provided functionalities, in this chapter, we are mainly concerned305

with how to provide telemonitoring and home support to improve users QoL.306

3.1 A Reference Scenario307

To better illustrate the objectives of the overall project and the urgent need for the308

approach presented in this chapter, let us illustrate a reference scenario.3309

Paul is a 60 years old man depressed about his recent stroke. Although Paul does310

not want to try new technologies, Dr. Jones suggests him to try to use a BNCI system311

at home, because he heard good things about the new BackHome system. Thus,312

Dr. Jones asks to Amanda a nurse with over 10 years experience helping people in313

managing care environments and tools to work with Jonas. At the beginning, she says314

that she does not want to. In fact, in the past, she had a bad experience mounting315

the cap, getting a new connection, and dealing with all the hassles of getting a316

BNCI to work. Dr. Jones asks her to try again and Amanda visits Pauls home. The317

first day, Amanda shows to Paul how to use the BackHome system and how it is318

easy to perform different tasks. The second day, Paul decides to try it and, thanks319

to the friendly support tools, Amanda is easily able to find all the solutions to the320

encountered troubles. In the next days, through the telemonitoring stations located321

at the hospital and at Pauls home, respectively, Dr. Jones is able to continuously322

verify the status of Paul and to suggest him new and personalized exercises to his323

rehabilitation therapy. In few weeks, Paul becomes more motivated, performs the324

rehabilitation exercises daily, joins a chess club, and starts to talk online to friends.325

Through BackHome system, Dr. Jones notes the progresses in Pauls daily activities326

and in his mood, and the corresponding general improved quality of life. Thus, he327

decides to assign other nurses to introduce BackHome to further patients.328

3.2 The BackHome Platform329

Before illustrating the proposed methodology to assess QoL through context aware-330

ness, let us introduce the BackHome platform, its main modules and the provided331

functionalities.332

3 Names have been changed for privacy reasons.


Au

tho

r P

roo

f

UN

CO

RR

EC

TE

D P

RO

OF

10 E. Vargiu et al.

B&

WIN

PRIN

T

Fig. 1 BackHome platform overview

The BackHome platform, depicted in Fig. 1, relies on two stations: (i) the therapist333

station and (ii) the user station. The former is responsible for the communication334

between the therapist and the user, as well as on handling Cognitive Rehabilitation335

(CR) tasks and QoL assessment. The latter is the main component which the user336

interacts with. It contains the modules responsible for the user interface, the execution337

of CR tasks, the control of the smart home and the other services, and the intelligence338

of the system, including the algorithm to assess the QoL.339

3.2.1 Therapist Station340

The therapist station is focused on offering information and services to the therapists341

via a usable and intuitive user interface (see Fig. 2). It is a Web application that allows342

the therapist to access the information of the patient independently of the platform343

and the device. This flexibility is important in order to get the maximum potential344

out of the telemonitoring because the therapist can be informed at any moment with345

any device that is connected to the Internet (PC, a smart phone or a tablet).346

Following a modular approach, the therapist station implements its functionalities347

in three main blocks, each one built upon a core module that provides cross-platform348

functionalities, such as user authentication, security, user and role management and349

platform configuration.350

The Cognitive Rehabilitation Module allows the therapists to manage CR tasks351

as well as remotely configure and program rehabilitation sessions for the users. This352


Au

tho

r P

roo

f

UN

CO

RR

EC

TE

D P

RO

OF


B&

WIN

PRIN

T

Fig. 2 A screen-shot of the therapist station interface

module also handles rehabilitation results and builds reports to allow therapist to353

follow-up the performed tasks. Also, alerts will be sent to the therapist in instance a354

user did not perform the assigned task.355

Similarly, the Quality of Life Module allows the therapist to remotely assess user’s356

QoL through predefined questionnaires that can be sent and programmed in this357

module. In particular, the therapist assigns and schedules QoL questionnaire(s) to358

the user. Alerts will be sent to the therapist in the event a user did not complete an359

assigned questionnaire. Moreover, this module receives from the user station the data360

concerning the questionnaire completed by the users and results coming from the361

automated QoL assessment system.362

Finally, the User Telemonitoring Module is in charge of handling and allowing363

direct communication between the therapist and a user. Communication are per-364

formed by relying on a teleconference system. Moreover, as a dispatcher, this module365

sends the right information on CR to the User Rehabilitation Module and on QoL to366

the Quality of Life Module, respectively.367

3.2.2 User Station368

The user station is the main component that the user interacts with. It contains the369

modules responsible for the user interface, the intelligence of the system (including370

the algorithms to assess the QoL and those to provide Context-Awareness), as well as371

to provide all the services and functionalities of BackHome. The user station will be372

completely integrated into the home of the user together with the assistive technology373

to enable execution and control of these functionalities.374

The BCI Block contains all the elements needed to allow the user to interact with375

the system and its services. It records the brain signals used to identify the selection376

made by the user. This is done using three different paradigms simultaneously (P300377


Au

tho

r P

roo

f

UN

CO

RR

EC

TE

D P

RO

OF

12 E. Vargiu et al.

[25], SSVEP [33] and SMR/MI [45]). At the same time the BCI block assesses378

the user’s fatigue state, which is used to switch the BackHome prototype into a379

stand-by mode in case the user is not attending or sleeping. A dedicated fusion380

process will ensure that only selections are conveyed to the attached applications381

and services. The corresponding output is then converted into their corresponding382

application and service commands. Subsequently, these will be transmitted to the383

different applications, services and BackHome systems using a dedicated network384

protocol.385

The AmI Block includes the elements for communicating with the BCI Block, the386

different services of the platform and the therapist station. This module also includes387

all the intelligence of the system devoted to process all the information from the user388

habits, the ambient, and the social interaction with the aim of helping the user to389

get the maximum potential of the system. The Communication Manager is the key390

module of the AmI Block. It implements the different interfaces interconnecting the391

AmI Block with the Therapist Station and the BCI Block. It also is the responsible to392

communicate the user’s action to the Service Manager, which is responsible for the393

execution of the actions by the smart home devices. This module implements several394

interfaces to connect all the devices and services supported by the platform. The395

Proactive Reasoning Engine constantly processes the actions of the user, the changes396

in the environment, and the services. It is aimed at understanding the context, detect397

habits, and predict situations that can help the user to better interact with the system398

and get more comfortable with it. In other words, it is in charge of processing data399

by relying on machine learning techniques aimed at learning from the user and the400

environment and adapting accordingly. The processed data will be used to change401

the context (by the Context Awareness Module) and/or to assess user’s QoL (by the402

Quality of Life Module).403

The User Station provides several services and applications:404

• Smart Home. The user can interact and control home devices, such as light, TV,405

and air conditioning. Moreover, environmental sensors allows the user to interact406

with the environment.407

• Cognitive Rehabilitation. It is the service that allows the interactions with the AmI408

Block to perform CR tasks.409

• Leisure. Through the BackHome platform, the user is able to interact with a suitable410

multimedia player. Moreover, s/he can use Brain Painting [34] to draw.411

• Communication. Through a suitable browser, the user can navigate the Internet412

and handling emails. Moreover, s/he can communicate and exchange information413

with the most popular social networks (i.e., Twitter and Facebook).414

4 Context-Aware Quality of Life Assessment415

As already said, this chapter aims to propose a general methodology to assess QoL416

by relying on context-aware techniques. The proposed methodology is currently417

adopted in the BackHome project. Nothing prevents to adopt it in a more general way418


Au

tho

r P

roo

f

UN

CO

RR

EC

TE

D P

RO

OF


in presence of persons to be remotely monitored or in case of disorder preventions.419

In fact, typically beneficiary will be older people or chronic patients.420

The underlying idea of the proposed methodology comes from a previous work421

[53]. To our best knowledge it is the first attempt to adopt context-aware techniques422

to assess QoL.423

4.1 The Proposed Telemonitoring System424

To monitor the QoL of disabled people, we propose a sensor-based Telemonitoring425

and Home Support System (TMHSS) aimed at helping the user to be more indepen-426

dent by providing smart home control. It also increases the eInclusion thanks to the427

possibility to perform Web browsing, use e-mail services, as well as interact with the428

most popular social networks.429

The sensor-based TMHSS is able to monitor the evolution of the user’s daily life430

activity at home, once discharged from the hospital [55], providing QoL automated431

assessment based on information gathering and data mining techniques [54]. Specif-432

ically, wearable sensors allow to monitor fatigue, spasticity, stress, and further user’s433

conditions. Environmental sensors are used to monitor—for instance—temperature434

and humidity, as well as the movements (motion sensors) and the physical position435

of the user (location sensors). Smart home devices enable physical autonomy of the436

user and help her/him carry out daily life activities. From the social perspective,437

an Internet-connected device allows the user to communicate with remote thera-438

pists, careers, relatives, and friends through Skype, email, or social networks (i.e.,439

Facebook and Twitter).440

The proposed sensor-based system acquires personalized information through441

data coming from: (i) a BNCI system4 that allows monitoring ElectroEncephalo-442

Gram (EEG), ElectroOculoGram (EOG), and ElectroMyoGram (EMG) signals; (ii)443

wearable, physiological, and biometric sensors, such as ElectroCardioGram (ECG)444

sensor, heart-rate sensor, respiration-rate sensor, Galvanic Skin Response (GSR)445

sensor, EMG switches, and inertial sensors (e.g., accelerometer, gyrocompass, and446

magnetometer); (iii) environmental sensors (i.e. gas, smoke, luminosity, humidity,447

and temperature sensors); (iv) smart home devices (e.g., home lights and TV); and448

(v) devices that allow interaction activities (i.e., a desktop PC).449

4.2 Quality of Life450

Starting from the standard questionnaires found in the literature, we propose a new451

Visual Analogue Scale (VAS) QoL questionnaire (see Fig. 3). The proposed ques-452

tionnaire is based on the standard EQ-5D-5L questionnaire, is designed to assess453

4 Currently, the EEG-P300-2D, a standard P300 control paradigm.


Au

tho

r P

roo

f

UN

CO

RR

EC

TE

D P

RO

OF

14 E. Vargiu et al.

B&

WIN

PRIN

T

Fig. 3 BackHome questionnaire

the key QoL features of an individual, which correspond with the main features454

that BackHome aims to monitor. In other words, we consider the user’s QoL as the455

conjunction of the following items: Mood, Health Status, Mobility, Self-care, Usual456

Activities, and Pain/Discomfort. According to [18], Table 1 shows the translation of457

the selected questionnaire into the ICF categories.458


Au

tho

r P

roo

f

UN

CO

RR

EC

TE

D P

RO

OF


Table 1 The translation of the items of the adopted questionnaire into the ICF categories

Questionnaire item ICF category

Mood b152—emotional functionsHealth status b130—energy and drive functions

b134—sleep functionsb730—muscle power functionsb735—muscle tone functionsb760—control of voluntary movement functions

Mobility d4—mobilityd450—walkingd498—mobility, other specified

Self-care d5—self-cared510—washing oneselfd540—toiletingd540—dressing

Usual activities d6409—doing housework, unspecifiedd7609—family relationships, unspecifiedd839—education, other specified and unspecifiedd8509—remunerative employment, unspecifiedd9209—recreation and leisure, unspecified

Pain/discomfort b152—emotional functionsb280—sensation of painb289—sensation of pain, other specified and unspecified

4.3 Context and Quality of Life459

Studying the different items that compose off-the-shelf QoL questionnaires as –for460

instance– the one presented in the previous section, some similarities with the concept461

of “context” can be noted. To highlight these similarities, let us consider the follow462

classification of QoL items:463

• User Information, information related to physical and mental health (e.g, mood464

and pain).465

• Interactions, environmental interaction (e.g., control over home environment) and466

social relationships (e.g., face-to-face communication and telecommunication).467

• Location, information related to user’s position as well as her/his movements (e.g.,468

mobility).469

• Daily activities, activities performed by the user (e.g., leisure activities).470

Although not explicitly shown, also the time is an important issue in compiling471

a questionnaire. In fact, it should modify the perception of the user’s surrounding472

influencing her/his status and, thus, the overall QoL. For instance, mood can change473

depending on the hour of the day, the season, and the number of sleeping hours.474

It is easy to note that the classification given above matches very well with the475

definition of “context”, given in [59] and recalled in Sect. 2.2. Thus, starting from that476


Au

tho

r P

roo

f

UN

CO

RR

EC

TE

D P

RO

OF

16 E. Vargiu et al.

definition, let us classify the QoL items according to the “five fundamental categories477

for context information”:478

• Individuality, it describes the state of the entity. In QoL, being the entity the479

individual, Individuality corresponds to the User Information category.480

• Relations, it describes the relations among different entities of a context-aware481

system. In QoL, it can be viewed as relations among entities (e.g., users), as well482

as “external” entities (e.g., caregivers and relatives). In a more broad view, it483

could also consider interaction with the environment. In other words, Relations484

corresponds to the Interactions category of the previous classification.485

• Location, it describes the position of an entity. In QoL, it is translated into the486

position of the individual and her/his ability in moving around.487

• Activity, it describes the activity corresponding to an entity in a context-aware488

scenario. In QoL, it covers all the daily life activities performed by the individual.489

Thus, it corresponds to the Daily Activities category.490

• Time, it describes the temporal dimension of the gathered information and it is491

really important in the classification of context [20]. In the case of QoL, “time”492

not only affects the context status of the individual’s surrounding, it also influences493

her/his physical status (e.g., the same fatigue value associated to daily activities494

has a different impact depending on the time in which it is gathered). As already495

said, Time does not have a direct correspondence with the QoL items. On the other496

hand, it can be considered as a “transversal” category that affects all the others.497

The correspondence between context and QoL assessment allows us to study how498

to automatically assess QoL by relying on context-aware techniques. In fact, those499

techniques have been proposed and used for recognizing activities and behavioural500

patterns [3, 36] or monitoring diet and exercise [16]. Similarly, we claim that context-501

aware techniques can be adopted to automatically assess QoL of individuals.502

Keeping in mind the above classifications, we can identify all the sensors involved503

in the process of gathering data to assess QoL:504

• Individuality505

– the BNCI system, which allows monitoring EEG, EOG, and EMG signals;506

– wearable, physiological, and biometric sensors, such as ECG sensor, heart-rate507

sensor, respiration-rate sensor, GSR sensor, EMG switches, and inertial sensors508

(e.g., accelerometer, gyrocompass, and magnetometer).509

• Relations510

– social networking (i.e., through Facebook and Twitter)511

– communications to the therapists (i.e., through the telemedicine platform).512

• Location513

– environmental sensors (e.g., temperature and humidity sensors);514

– inertial, location, and motion sensors.


Au

tho

r P

roo

f

UN

CO

RR

EC

TE

D P

RO

OF


B&

WIN

PRIN

T

Fig. 4 Gathered data in BackHome and their relation with the proposed categories

• Activity515

– smart home devices (e.g., wheelchairs, lights, TVs, doors, windows and shut-516

ters);517

– devices that allow interaction activities (e.g., a desktop PC);518

– devices to perform rehabilitation tasks (e.g., a robot).519

Figure 4 shows the complete set of information available highlighting the category520

to which each input belongs to: pentagons correspond to Individuality; rectangles to521

Relations; hexagons concern with Location; and circles with Activity.522

4.4 The Approach523

Personalised information will be captured through the combination of data coming524

from the sensor-based system. This information will be fused with that gathered525

when the user is interacting with the BackHome platform and with questionnaires,526

if needed. The data will be used to inform the system of the users’ behaviours, social527

autonomy, and to other support tasks. In particular, two kinds of data are considered:528

monitorable and inferable. Monitorable data can be gathered from the wearable,529

home automation, and environmental sensors, as well as the BNCI system (i.e.,530

without relying on direct input from the user). For example, this kind of data provides531


Au

tho

r P

roo

f

UN

CO

RR

EC

TE

D P

RO

OF

18 E. Vargiu et al.

the answer to the item on “Mobility” (Today, my ability to move about was...). On532

the other hand, all data inferred by analysing data retrieved by the system (e.g., by533

considering activities performed by the user while interacting with a social network)534

belong to the latter category. This kind of data allows, for instance, answering the535

item on “Mood” (Today, my overall mood was...).536

Let us note that this does not imply that monitorable and inferable data are neces-537

sarily monitored or inferred. In particular, in BackHome we decided to not monitor538

nor infer some data (such as, those related to Self-care and sleeping activities), due539

to privacy issues. Moreover, users can decide to switch off the monitoring of any540

descriptor.541

In the following, we briefly describe how each monitored data can be gathered to542

assess the items of the given questionnaire.543

Monitorable data544

Health Status: Through the adoption of wearable, physiological, and biometric545

sensors, the system is able to monitor improvement and/or worsening of the health546

status of the user.547

Mobility: Through the adoption of environmental location sensors, the system is548

able to know the position of the user, time after time. It is worth pointing out that,549

in BackHome, users are typically on a wheelchair, thus the walking activity is not of550

interest here. To detect the position of the wheelchair and its movements, RFID tags551

could be embedded into the wheelchair together with following sensors.552

Usual Activities: Being human-computer-interaction made through a BNCI sys-553

tem, it is possible to monitor all the activities performed by the user on the PC and554

while interacting with smart home control and communication devices. In other555

words, the system is able, through the BNCI system, to know which action is556

performed, such as home environment interactions, face-to-face communications,557

telecommunications, and leisure activities. Moreover, the activities performed on558

further devices that allow some kind of interaction and stimulation activities (e.g.,559

devices to game, hear music, perform painting activities and/or further leisure activ-560

ities) will be stored to further studies on the user’s interaction and leisure activities.561

Inferable data562

Mood: Changes observed in habits of daily life activities can be studied to assess563

the mood. The degree of overall satisfaction can be also inferred by analysing data on564

fatigue, spasticity, stress, and further users conditions retrieved by the BNCI system565

and the other wearable sensors. Moreover, analogously to pain and discomfort, anx-566

iety and depression can be inferred by the system by adopting suitable text mining567

algorithms on the performed social activities.568

Usual Activities: The user can interact with her/his family and friends through the569

support of a communication system (e.g., Skype) or social network (e.g., Facebook570

and Twitter). Thus, suitable text mining algorithms can be adopted to infer the family571

and friend relationships.572

Pain/Discomfort: Text mining algorithms, applied to social networking and com-573

munication activities, will be adopted to assess the degree of pain or discomfort. Of574

course, privacy and technological considerations will be taken into account to define575

the scope of those analyses.576


Au

tho

r P

roo

f

UN

CO

RR

EC

TE

D P

RO

OF


4.5 Preliminary Results577

Among all the features that the TMHSS will be able to assess, we started from578

monitorable data. In particular, our first step was to monitor Usual Activities through579

the support of the BNCIs. Subsequently, we defined how to monitor Mobility, the580

corresponding experiments are currently running.581

As for Usual Activities, we performed a study similar to the one we proposed in [8]582

and used in the BrainAble project5 [17]. In particular, we rely on a C4.5 decision-tree583

classifier to study the surrounding environment and user’s habits. According to our584

previous work, we first tested the classifier to learn user’s preferences at the different585

moments of the day (e.g., morning, afternoon, night). The classifier receives as input586

the actual state of the environment, described as temporal information, together with587

the set of states of all the devices the user is able to interact with through the BNCI588

system (e.g., doors, home lights, and TV). The output of the classifier consists of a589

set of user’s preferences to be suggested in real time. Due to the lack of real data,6590

according to [1], we generated synthetic data. Experiments have been performed by591

running 10-fold cross-validation, in which the dataset has been randomly partitioned592

into 10 subsets. For each fold, one subset has been chosen to validate the model, and593

the other 9 to train the classifier. Results have then been averaged to produce a single594

estimation. Results show an accuracy of about 99 % (see Fig. 5). We are currently595

B&

WIN

PRIN

T

Fig. 5 Accuracy of the adopted classifier

5 www.brainable.org6 The first prototype of the system has been installed at the end-user facilities at the beginning of2013.


Au

tho

r P

roo

f

www.brainable.org

UN

CO

RR

EC

TE

D P

RO

OF

20 E. Vargiu et al.

setting up a new classifier instance able also to recognise the performed activities596

(instead of suggesting ones) to monitor the overall user’s behaviour and its trend.597

As for Mobility, through the adoption of inertial, location, and motion sensors,598

the system is able to know the position of the user, time after time. All information599

about location, performed movements, covered distance, visited rooms, time spent600

on the bed (and thus on the wheelchair) are used as classification features to build a601

multi-class k-NN. The considered classes concern the user’s satisfaction in her/his602

mobility ability (from “Very Bad” to “Very Good”) and the training set is built directly603

asking users to assess their level of satisfaction. Once the system has been trained,604

we are able to infer user’s satisfaction, to study the behaviour trend, and to assess605

the improvement/worsening of the user’s QoL. Results will be evaluated in term of606

classical information theory measures, i.e., precision, recall, and F1.607

5 Conclusions608

A methodology to telemonitor quality of life based on the awareness of user context609

(fusion and processing of heterogeneous data coming from sensors, services and user610

interaction) has been proposed.611

This methodology has been devised in the framework of a particular application,612

the BackHome project and BackHome platform. Nonetheless, it holds a generic613

approach, able to be adapted to other telemedicine and teleassistance applications.614

The problem we are tackling here is very ambitious. In order to properly validate615

and enhance this methodology we will need to get big and varied amounts of data,616

from an extensive sample of users, within a wide range of conditions and environ-617

ments, along representative periods of time. This work will be continued during the618

duration of the BackHome project and others projects to come.619

Acknowledgments The research leading to these results has received funding from the European620

Community’s, Seventh Framework Programme FP7/2007-2013, BackHome project grant agree-621

ment n. 288566.622

References623

1. Armac, I., Retkowitz, D.: Simulation of smart environments. In: Pervasive Services, IEEE624

International Conference on, pp. 322–331 (2007)625

2. Artinian, N.: Effects of home telemonitoring and community-based monitoring on blood pres-626

sure control in urban African Americans: a pilot study. Heart Lung 30, 191–199 (2001)627

3. Barger, T.S., Brown, D.E., Alwan, M.: Health-status monitoring through analysis of behavioral628

patterns. IEEE Trans. Syst. Man Cybern. Part A Syst. Hum. 35(1), 22–27 (2005)629

4. Bhattacharyya, S.: Context aware health care application. Int. J. Adv. Technol. 2(3), 461–470630

(2011)631


Au

tho

r P

roo

f

UN

CO

RR

EC

TE

D P

RO

OF


5. Birbaumer, N., Ghanayim, N., Hinterberger, T., Iversen, I., Kotchoubey, B., Kubler, A., Perel-632

mouter, J., Taub, E., Flor, H.: A spelling device for the paralysed. Nature 398(6725), 297–298633

(1999)634

6. Byun, H.E., Cheverst, K.: Exploiting user models and context-awareness to support personal635

daily activities. In: Workshop in UM2001 on User Modeling for Context-Aware Applications636

(2001)637

7. Carneiro, D., Costa, R., Novais, P., Machado, J., Neves, J.: Simulating and monitoring ambient638

assisted living. In: Proceeding of ESM (2008)639

8. Casale, P., Fernández, J.M., Rafael, X., Torrellas, S., Ratsgoo, M., Miralles, F.: Enhanching640

user experience with brain neural computer interfaces in smart home environments. In: 8th641

IEEE International Conference of Intelligent, Environments 2012, INTENV12 (2012)642

9. Ceccaroni, L., Subirats, L.: Interoperable knowledge representation in clinical decision support643

systems for rehabilitation. Int. J. Appl. Comput. Math. 11(2), 303–316 (2012)644

10. Cieza, A., Geyh, S., Chatterji, S., Kostanjsek, N., Üstün, B., Stucki, G.: Icf linking rules: an645

update based on lessons learned. Rehabil. Med. 37, 212–218 (2005)646

11. Corchado, J., Bajo, J., Tapia, D., Abraham, A.: Using heterogeneous wireless sensor networks647

in a telemonitoring system for healthcare. IEEE Trans. Inf. Technol. Biomed. 14(2), 234–240648

(2010)649

12. Cordisco, M., Benjaminovitz, A., Hammond, K., Mancini, D.: Use of telemonitoring to decrease650

the rate of hospitalization in patients with severe congestive heart failure. Am. J. Cardiol. 84(7),651

860–862 (1999)652

13. Daly, J., Armstrong, E., Miralles, F., Vargiu, E., Müller-Putz, G., Hintermller, C., Guger, C.,653

Kuebler, A., Martin, S.: Backhome: Brain-neural-computer interfaces on track to home. In:654

RAatE 2012—Recent Advances in Assistive Technology and Engineering (2012)655

14. Dey, A.K.: Understanding and using context. Pers. Ubiquitous Comput. 5(1), 4–7 (2001)656

15. Edlinger, G., Holzner, C., Guger, C.: A hybrid brain-computer interface for smart home con-657

trol. In: Proceedings of the 14th International Conference on Human-Computer Interaction:658

Interaction Techniques and Environments—Volume Part II, HCII’11, pp. 417–425. Springer,659

Berlin (2011)660

16. Farringdon, J., Nashold, S.: Continuous body monitoring. In: Ambient Intelligence for Scien-661

tific Discovery, pp. 202–223. Springer, Berlin (2005)662

17. Fernández, J.M., Dauwalder, S., Torrellas, S., Faller, J., Scherer, R., Omedas, P., Verschure,663

P., Espinosa, A., Guger, C., Carmichael, C., Costa, U., Opisso, E., Tormos, J., Miralles, F.:664

Connecting the disabled to their physical and social world: The BrainAble experience. In: TOBI665

Workshop IV Practical Brain-Computer Interfaces for End-Users: Progress and Challenges666

(2013)667

18. Geyh, S., Cieza, A., Kollerits, B., Grimby, G., Stucki, G.: Content comparison of health-related668

quality of life measures used in stroke based on the international classification of functioning,669

disability and health (icf): a systematic review. Qual. Life Res. 16(5), 833–851 (2007)670

19. Gotay, C., Moore, T.: Assessing quality of life in head and neck cancer. Qual. Life Res. 1, 5–17671

(1992)672

20. Gross, T., Specht, M.: Awareness in context-aware information systems. In: Mensch and Com-673

puter 2001, pp. 173–182. Springer, New York (2001)674

21. Hays, R., Morales, L., RAND Health, P., RAND, C.: The RAND-36 Measure of Health-related675

Quality of Life. RAND Health reprint. RAND (2001)676

22. Hitzler, P., Krötzsch, M., Parsia, B., Patel-Schneider, P.F., Rudolph, S. (eds.): OWL 2 Web677

Ontology Language: Primer. W3C Recommendation (2009)678

23. Holzner, C., Schaffelhofer, S., Guger, C., Groenegress, C., Edlinger, G., Slater, M.: Using a679

p300 brain-computer interface for smart home control. In: World Congress (2009)680

24. Intille, S.S., Kaushik, P., Rockinson, R.: Deploying Context-Aware Health Technology at681

Home: Human-Centric Challenges. In: Human-Centric Interfaces for Ambient Intelligence682

(2009)683

25. Jin, J., Allison, B.Z., Sellers, E.W., Brunner, C., Horki, P., Wang, X., Neuper, C.: Optimized684

stimulus presentation patterns for an event-related potential eeg-based brain-computer inter-685

face. Med. Biol. Eng. Comput. 49(2), 181–191 (2011)686


Au

tho

r P

roo

f

UN

CO

RR

EC

TE

D P

RO

OF

22 E. Vargiu et al.

26. Lerer, L.: The healthcare 2020 platform: The e-health consumer. Ph.D. thesis, INSEAD (2000)687

27. Marilyn J. Field Editor; Committee on Evaluating Clinical Applications of Telemedicine,688

I.o.M.: Telemedicine: A Guide to Assessing Telecommunications for Health Care. The National689

Academies Press, Washington (1996)690

28. Martn-Lesende, I., Orruño, E., Cairo, C., Bilbao, A., Asua, J., Romo, M., Vergara, I., Bayn,691

J., Abad, R., Reviriego, E., Larrañaga, J.: Assessment of a primary care-based telemonitoring692

intervention for home care patients with heart failure and chronic lung disease. The TELBIL693

study. BMC Health Serv. Res. 11(56), (2011)694

29. Mason, S.G., Bashashati, A., Fatourechi, M., Navarro, K.F., Birch, G.E.: A comprehensive695

survey of brain interface technology designs. Ann. Biomed. Eng. 35, 137–169 (2007)696

30. Meystre, S.: The current state of telemonitoring: a comment on the literature. Telemed. J. E697

Health 11(1), 63–69 (2005)698

31. Mitchell, M., Meyers, C., Wang, A., Tyson, G.: Contextprovider: context awareness for med-699

ical monitoring applications. In: Conference Proceeding IEEE Engineering in Medicine and700

Bioliogy Society (2011)701

32. Müller, G., Neuper, C., Pfurtscheller, G.: Implementation of a telemonitoring system for the702

control of an EEG-based brain-computer interface. IEEE Trans. Neural Syst. Rehabil. Eng.703

11(1), 54–59 (2003)704

33. Müller-Putz, G.R., Pfurtscheller, G.: Control of an electrical prosthesis with an ssvep-based705

bci. IEEE Trans. Biomed. Eng. 55(1), 361–364 (2008)706

34. Münssinger, J., Halder, S., Kleih, S., Furdea, A., Raco, V., Hösle, A., Kübler, A.: Brain painting:707

first evaluation of a new braincomputer interface application with als-patients and healthy708

volunteers. Front Neurosci. 4, 182. doi: 10.3389/fnins.2010.00182 (2010)709

35. Murphy, B., Herrman, H., Hawthorne, G., Pinzone, T., Evert, H.: Australian WHOQoL instru-710

ments: user’s manual and interpretation guide. Australian WHOQoL Field Study Centre. Mel-711

bourne, Australia (2000)712

36. Nambu, M., Nakajima, K., Noshiro, M., Tamura, T.: An algorithm for the automatic detection713

of health conditions. Eng. Med. Biol. Mag. IEEE 24(4), 38–42 (2005)714

37. Naughton, M., Shumaker, S., Anderson, R., Czajkowski, S.: Psychological aspects of healthre-715

lated quality of life measurement: tests and scales. Qual. Life Pharmaco Econ. Clin. Trials 15,716

117–131 (1996)717

38. Navarro, A., Ceccaroni, L., Velickovski, F., Torrellas, S., Miralles, F., Allison, B., Scherer, R.,718

Faller, J.: Context-awareness as an enhancement of brain-computer interfaces. In: III Interna-719

tional Workshop on Ambient Assisted Living IWAAL 2011 (2011)720

39. Neuper, C., Pfurtscheller, G.: Neurofeedback Training for BCI Control Brain-Computer Inter-721

faces. The Frontiers Collection, Chap. 4, pp. 65–78. Springer, Berlin (2010)722

40. O’Sullivan, S., Schmitz, T.: Physical Rehabilitation. Information and Interdisciplinary Subjects723

Series. F.A. Davis, G— Reference (2007)724

41. Patrick, D., Erickson, P.: Health Status and Health Policy: Quality of Life in Health Care725

Evaluation and Resource Allocation. No. p. 314 in Health Status and Health Policy: Quality of726

Life in Health Care Evaluation and Resource Allocation. Oxford University Press, New York727

(1993).http://books.google.es/books?id=IjgQAQAAMAAJ728

42. Pfurtscheller, G., Allison, B.Z., Brunner, C., Bauernfeind, G., Solis-Escalante, T., Scherer, R.,729

Zander, T.O., Mueller-Putz, G., Neuper, C., Birbaumer, N.: The hybrid BCI. Frontiers Neurosci.730

4 (2010)731

43. Pfurtscheller, G., Müller-Putz, G.R., Scherer, R., Neuper, C.: Rehabilitation with brain-732

computer interface systems. IEEE Comput. 41(10), 58–65 (2008)733

44. Rand, C., Health Sciences, P., Hays, R., Sherbourne, C., Mazel, R.: RAND 36-item Health734

Survey 1.0: RAND Health Sciences Program. RAND reprints. RAND (1992)http://books.735

google.es/books?id=FUkKHAAACAAJ736

45. Sannelli, C., Dickhaus, T., Halder, S., Hammer, E.M., Müller, K.R., Blankertz, B.: On optimal737

channel configurations for smr-based brain-computer interfaces. Brain topogr. 23(2), 186–193738

(2010)739


Au

tho

r P

roo

f

http://books.google.es/books?id=IjgQAQAAMAAJ

http://books.google.es/books?id=FUkKHAAACAAJ

http://books.google.es/books?id=FUkKHAAACAAJ

UN

CO

RR

EC

TE

D P

RO

OF


46. Barro, S., Castro, D., Fernndez-Delgado, M., Fraga, S., Lama, M., Rodrguez, J.M., Vila, J.A.:740

Intelligent telemonitoring of critical-care patients. IEEE Eng. Med. Biol. Mag. 18, 80–88 (1999)741

47. Schilit, B., Adams, N., Want, R.: Context-aware computing applications. In: Mobile Computing742

Systems and Applications, 1994. WMCSA 1994. First Workshop on, pp. 85–90. IEEE (1994)743

48. Skevington, S.M., Lotfy, M., O’Connell, K.A., WHOQOL Group: The World Health Organiza-744

tion’s WHOQOL-BREF quality of life assessment: psychometric properties and results of the745

international field trial. A report from the WHOQOL group. Qual. Life Res. Int. J. Qual. Life A.746

Treat. Care Rehabil. 13(2), 299–310 (2004). http://view.ncbi.nlm.nih.gov/pubmed/15085902747

49. Subirats, L., Ceccaroni, L., Miralles, F.: Knowledge representation for prognosis of health748

status in rehabilitation. Future Internet 4(3), 762–775 (2012)749

50. Sutherland, H., Till, J.: Quality of life assessments and levels of decision making: differentiating750

objectives. Qual. Life Res. 2(4), 297–303 (1993). http://www.biomedsearch.com/nih/Quality-751

life-assessments-levels-decision/8220364.html752

51. The Euroqol Group: Euroqol a facility for the measurement of health-related quality of life.753

Health Policy 16, 199–208 (1990)754

52. Tonin, L., Leeb, R., Tavella, M., Perdikis, S., Millán, J.: A bci-driven telepresence robot. Int.755

J. Bioelectromagn. 13(3), 125–126 (2011)756

53. Vargiu, E., Ceccaroni, L., Subirats, L., Martin, S., Miralles, F.: User profiling of people with757

disabilities - a proposal to pervasively assess quality of life. In: 5th International Conference758

on Agents and Artificial Intelligence (2013)759

54. Vargiu, E., Fernández, J.M., Torrellas, S., Dauwalder, S., Solà, M., Miralles, F.: A sensor-760

based telemonitoring and home support system to improve quality of life through bnci. In:761

12th European AAATE Conference (2013)762

55. Vargiu, E., Miralles, F., Martin, S., Markey, D.: BackHome: Assisting and telemonitoring people763

with disabilities. In: RAatE 2012—Recent Advances in Assistive Technology and Engineering764

(2012)765

56. Vincent, J., Cavitt, D., Karpawich, P.: Diagnostic and cost effectiveness of telemonitoring the766

pediatric pacemaker patient. Pediatr. Cardiol. 18(2), 86–90 (1997)767

57. Ware, J., Kosinski, M., Dewey, J.: How to Score Version 2 of the SF-36 Health Survey: Standars768

and Acute Forms. QualityMetric (2001)769

58. WHO: Global Age-Friendly Cities: A Guide. Nonserial Publication. World Health Organization770

(2007). http://books.google.es/books?id=4uWtQy6rGywC771

59. Zimmermann, A., Lorenz, A., Oppermann, R.: An operational definition of context. In: Mod-772

eling and using context, pp. 558–571. Springer, Berlin (2007)773


Au

tho

r P

roo

f

http://view.ncbi.nlm.nih.gov/pubmed/15085902

http://www.biomedsearch.com/nih/Quality-life-assessments-levels-decision/8220364.html

http://www.biomedsearch.com/nih/Quality-life-assessments-levels-decision/8220364.html

http://books.google.es/books?id=4uWtQy6rGywC

UN

CO

RR

EC

TE

D P

RO

OF

Author Queries

Chapter 1

Query Refs. Details Required Author’s response

AQ1 As keywords are mandatory for this chapter, please provide 3-6keywords.

AQ2 Please confirm the corresponding author is correctly identified andamend if necessary.

AQ3 Please provide page range for Ref. [28, 42].

AQ4 Please provide editor group and for Refs. [16, 20, 59].

Au

tho

r P

roo

f

MARKED PROOF

Please correct and return this set

Instruction to printer

Leave unchanged under matter to remain

through single character, rule or underline

New matter followed by

or

or

or

or

or

or

or

or

or

and/or

and/or

e.g.

e.g.

under character

over character

new character

new characters

through all characters to be deleted

through letter or

through characters

under matter to be changed





Encircle matter to be changed

(As above)

(As above)

(As above)

(As above)

(As above)

(As above)

(As above)

(As above)

linking characters

through character or

where required

between characters or

words affected

through character or

where required

or

indicated in the margin

Delete

Substitute character or

substitute part of one or

more word(s)Change to italics

Change to capitals

Change to small capitals

Change to bold type

Change to bold italic

Change to lower case

Change italic to upright type

Change bold to non-bold type

Insert ‘superior’ character

Insert ‘inferior’ character

Insert full stop

Insert comma

Insert single quotation marks

Insert double quotation marks

Insert hyphen

Start new paragraph

No new paragraph

Transpose

Close up

Insert or substitute space

between characters or words

Reduce space betweencharacters or words

Insert in text the matter

Textual mark Marginal mark

Please use the proof correction marks shown below for all alterations and corrections. If you

in dark ink and are made well within the page margins.

wish to return your proof by fax you should ensure that all amendments are written clearly

context-aware based quality of life telemonitoring

Documents