INTRODUCTION

Demographic ageing is the term given to theanticipated huge increase in the number of olderpeople in our society over subsequent genera-tions. It is accepted that our health services can-not continue to provide hospital-based care forthis growing cohort of older people. Increasingnumbers of people will experience a new,evolved health service provision in their lateryears, which will be technologically-driven andwill provide care services or access to care ser-vices in the person’s own home. Such servicescould include, for example, activity remindersfor people with mild dementia, or medicationreminders for people complying with complex

medication regimes.These new ambient assisted living (AAL) ser-

vices, underpinned by home-based assistive tech-nologies offer data streams that will provide richsources of useful knowledge about the behaviorand wellbeing of people accessing care, eithersingly or in aggregated form. The deployedInformation and Communications Technology(ICT) also facilitates new modes of interactionbetween the person at home, their family, carers,and other healthcare professionals involved.

The AAL services that provide support forpeople to remain in their homes are increasinglybeing used in healthcare systems around theworld. Typically, these AAL services offer addi-tional information though activity-sequence-awareness, location-awareness, presence-awareness, and context-awareness capabilities,arising from the pervasive use of telecommunica-tions devices including processors, sensors, andactuators in the home of the person receivingcare. There are many AAL endpoints now utiliz-ing and feeding information to converged net-works and the resulting explosion in data andinformation has resulted in a new problem.These new converged AAL services struggle toconvey meaningful and timely information todivergent groups of end users, with differentinformation needs.

In this article AAL technology and servicesare explained, before examples of the services incommercial systems and in academic researchare explored. Visualization in AAL services isthen described from the needs perspective of aperson availing of the support — the care recipi-ent — explaining how visualization must conveyinformation across location and time. In order toillustrate the challenges for inclusive design, cur-rent research on night-time AAL services forpeople with dementia is then described in a casestudy.

DEFINING AAL TECHNOLOGY AND

IEEE Communications Magazine • January 20112 0163-6804/11/$25.00 © 2011 IEEE

ABSTRACT

Ambient assisted living services that providesupport for people to remain in their homes areincreasingly being used in healthcare systemsaround the world. Typically, these ambientassisted living services provide additional infor-mation though location-awareness, presence-awareness, and context-awareness capabilities,arising from the prolific use of telecommunica-tions devices including sensors and actuators inthe home of the person receiving care. In addi-tion there is a need to provide abstract informa-tion, in context, to local and remotestakeholders. There are many different viewingoptions utilizing converged networks and theresulting explosion in data and information hasresulted in a new problem, as these new ambientassisted living services struggle to convey mean-ingful information to different groups of endusers. The article discusses visualization of datafrom the perspective of the needs of the differ-ing end user groups, and discusses how algo-rithms are required to contextualize and conveysinformation across location and time. In order toillustrate the issues, current work on night-timeAAL services for people with dementia isdescribed.

NEW CONVERGED TELECOMMUNICATIONAPPLICATIONS FOR THE END USER

Maurice Mulvenna, William Carswell, Paul McCullagh, Juan Carlos Augusto, and Huiru Zheng, University of Ulster

Paul Jeffers, Fold Housing AssociationHaiying Wang and Suzanne Martin, University of Ulster

Visualization of Data for Ambient Assisted Living Services

MULVENNA LAYOUT 12/3/10 8:34 PM Page 2

IEEE Communications Magazine • January 2011 3

SERVICES

Assisted living is the term given to the provisionof care to people either in their own homes or insupported housing, underpinned by technology.The provision of care, augmented by assisted liv-ing technologies, is growing because of theincreasing demand and also because of thematuring of many of the underlying technologiesthat make assisted living possible. In parallelwith the development of assisted living,researchers in computing have been exploringthe emerging area of ambient intelligence whichapplies automated reasoning and other artificialintelligence techniques to the understanding ofthe behavior of people in their environments.

Ambient intelligence has evolved with greatpace over the past ten years, from the earlybeginnings when the European CommissionISTAG group presented their vision for ambientintelligence [1] to research on the many applica-tions. Ambient intelligence applications havebecome more complex and are now integratedinto many other systems in the home, medicaland occupational environments. Ambient intelli-gence based systems provide feedback to usersand carry out specific actions based on observedpatterns and pre-programmed algorithms. Somesystems are also aware of their surroundings andcan function independently, offering capabilitiesincluding “sensitive, responsive, adaptive, trans-parent, ubiquitous, and intelligent” [2].

These two areas of assisted living and ambi-ent intelligence are converging towards a newparadigm in social computing called AmbientAssisted Living (AAL). AAL services have bene-fited from advances in sensor technology, hard-ware, software, and communication paradigms tosuch an extent that AAL services have gainedmarket penetration into the home, work andhealth environments. AAL may be implementedto either replace or complement the care provid-ed by the carer. As technology becomes increas-ingly mobile, ubiquitous and pervasive, it is of

course likely that the wider population willbecome beneficiaries of AAL and may lead anincreasingly technology-augmented lifestyle.

AAL technologies have been outlined astechnologies that may help to extend the timethat older people can live at home by “increasingtheir autonomy and assisting them in carryingout activities of daily life” [3]. The services thatAAL technologies may use include functional,activity, cognitive, intellectual, and sensory sup-port. Examples of functions that the AAL tech-nologies may provide include alarms to detectdangerous situations that are a threat to theuser’s health and safety, monitoring and continu-ously checking the health and well-being of theuser, and the use of interactive and virtual ser-vices to help support the user. AAL technologiesmay also be used for communication, enablingthe user to keep in touch with family, friends,and carers, and for example, in support of remi-niscing.

VISUALIZATION OF AAL SERVICEDATA: USERS, ROLES, AND

EXAMPLES OF USE

AAL services have to support very differentkinds of technologies encompassing sensors,actuators, communication hubs and interfaces.There are a number of different classes of usersof the services. In essence, AAL services utilizedata and information from these devices usingdifferent protocols and orchestrate this informa-tion for the different users. The primary user ofAAL services is the care recipient, but there areother important users including on-site formalcarers, remote carers or monitors of AAL ser-vices, informal carers (including family, neigh-bors and friends) and those in charge ofmaintaining the quality of service in a technicalservice provision. Each of these classes of usershas a role or roles to play in AAL service provi-

Table 1. AAL service users and their roles.

User Role

Care recipient User of the services; needs to be able to interact with the services to gain assistance to provide feedbackwhere needed.

Formal carers on-site Interacts with the services in tandem with the care recipient to assess progress in care regime; interacts toadd information into system, for example, to add medication reminders.

Formal carers off-site As above for formal carers on-site but may also interact remotely to update information and interact withservices that measure or provide care to more than one home.

Telecare/Telehealthremote monitors off-site

Managing remote care provision; providing care triage in decision support for interventions; monitoringrelatively large numbers of care recipients.

Informal carers on-site Accessing information on health and wellbeing of care recipient; working with care recipient to understandhealth and wellbeing of care recipient and communicate to care recipient.

Informal carers off-site Remote access to monitor key information on care recipient, for example, family member concerned aboutquality of life of care recipient.

Technical maintenance Deploying tools to assess quality of data being gathered, decision support, and reporting on metrics fromhomes of care recipients; detection of errors.

MULVENNA LAYOUT 12/3/10 8:34 PM Page 3

IEEE Communications Magazine • January 20114

sion and its use (Table 1).AAL services have evolved from relatively

simple telecare services such as emergency fallalarm provision into more sophisticated tele-health services supporting people with long-termchronic health conditions such as Alzheimer’sdisease, in the assessment of their symptoms. Inthis evolution, the data generated has increasedin volume and complexity. The task of interpret-ing the meaning in the data relating to the well-being and health of the care recipient has alsobecome increasingly more complex and thediverse range of types of user and their overlap-ping roles in AAL service provision and efficacyhas increased markedly.

In determining appropriate AAL services,understanding the behaviour of the care recipi-ents in terms of their Activities of Daily Living(ADL) is of particular value. For example, prob-abilistic models have been used to manageuncertainty and incompleteness of data as theADLs are undertaken. ADLs comprise commonactivities such as making tea, using the tele-phone, etc. Conveying the information throughvisualization of such activities is much easier fora carer to interpret and comprehend comparedto data-rich, lower-level data about movementwithin a room, for example. However, incorrectADL identification has the potential to intro-duce more significant errors which could beharmful or even life threatening for the carerecipient.

There is a need for the AAL services to com-municate vitally important information in an

easy-to-understand manner to all users whilemaintaining the privacy of the care recipient andtheir informal carers. The manner in which theseAAL services communicate information drawsheavily on visualization techniques. The mostimportant user, the recipient of AAL care ser-vices, is more likely to be someone who is notfamiliar with ICT or computers in general andthis is a major complicating factor in the success-ful provision of visualization of data and infor-mation in AAL services.

Data collected in AAL services can includemovement information, used to:• Alert an emergency incident such as a fall• Alert for unexpected behaviors, such as an

older person going out late at night• Remind and assist daily living activities,

such as doing physical activities and takingmedicationsThe AAL services store personal activity pro-

files over periods of time. The trend of userbehavior and the pattern of user activity eventsprovide rich information in the analysis of carerecipients’ behavior patterns.

Visualization of information for AAL systemsneeds adaptive interfaces that target specificneeds and preferences of the recipients of careservices. We may refer to this as user-awareness,but overall this type of knowledge that the sys-tem uses to deliver a more useful service can beconsidered part of the most widely standardterm of context-awareness. For example, for theperson being cared for different reminders maybe suitable in the morning (e.g., “time to takeyour medicine,” “you did not have breakfastyet”) than in the afternoon (e.g., “you did nothave lunch yet”). Day and night require differenttypes of content given that the person it is deliv-ered to may have different level of consciousnessand alertness. It is also expected that the mes-sages to be delivered during the night will befocused on sustaining a healthy sleeping patternwhilst during the day it could refer to manyother daily activities. The place of the housewhere a reminder has to be delivered can makea difference as well (e.g., kitchen or living roomdisplays may be assumed to be used with normallevels of lighting whilst those in the bedroomwhich may be mostly used with low lighting ordarkness).

COMMERCIAL EXAMPLESThere are several commercial products providingAAL services that include a visual elementthrough text-based, colored tables and charts.Each vendor tends to differ in terms of how itvisually represents data. While some systems areable to push pre-structured alerts to carers whencertain events are triggered, the majority of visu-al representation of data is provided in responseto user-driven queries from the carers ratherthan automatically being streamed to carers inreal time.

An example of an advanced system is the useof the ActivPal1 system in ambulatory monitor-ing, which has proved successful, in manyrespects due to powerful visualization software.This provides a concise easy-to-interpret graphi-cal representation of the motion activity of aperson for a day classified as sitting, standing or

Figure 1. Illustration of one-day activity profile recorded from a 72 year-oldsubject using ActivPal wearable sensor (PAL technologies, 2010). The record-ing covers his walking day of just over 14 hours.

1 http://www.paltechnolo-gies.com/

2 http://www.tunstall.co.uk/

MULVENNA LAYOUT 12/3/10 8:34 PM Page 4

IEEE Communications Magazine • January 2011 5

walking. Meaningful feedback of the user pro-files has proven to be an important tool in assess-ing users’ behavior. Figure 1 illustrates arecording of a user’s 14 hours activity using anActivPal sensor. The profile displays the activitypattern during a day. It shows that the user wasinactive most of the time and was only active at9 am, 10 am, and 4 pm. Using this daily profile,carers would be able to spot the periods wherethe care recipient can improve their activitylevel. Comparing the daily profile to the user’sweekly profile, many questions can be answeredor explored, for example, if the user hadincreased/decreased his activity level comparedto the previous days? Did the user change theirlife style this day (for example, the user didn’tget up for lunch)? If yes, can we infer the rea-son?

Tunstall’s ADLife2 service provides a userinterface for representing client data, which isintended for use by healthcare professionals.The setup provides the user with information onindividual care recipients in the form of a tablelisting the total activations of all the sensoryequipment over a day, week and month. Eachcell is color coded to represent activity based ona standard deviation from the previous week’spattern, with green showing under use and redover usage. Each cumulative activation numbercell and date cell is hyperlinked to the completedata sets for that period, which are returned inthe form of bar charts. These indicate periods ofactivations over time and the total number ofactivations for each sensor device.

The QuietCare system3 caters for formal andinformal carer needs. The home page is orga-nized around activities rather than specific sen-sors providing sensor information in a moremeaningful structured output. A simple trafficlight indicator is used as the status display foreach activity. In addition there is meaningful textrelated to each area. Contact information is also

provided for alarm conditions. If the user wantsto view specific trend details they can select theview buttons, which provides information foreach activity on the last week, along with a textu-al description. If the user requests more detaileddata they can select to view it from a sub-menu.While both these examples provide visual repre-sentation of data through imagery or well-struc-tured text, there is variation as to how servicedata is presented.

ACADEMIC EXAMPLESIn academia, researchers have focused on lower-ing the cognitive processing required in order tointerpret the data being visualized. For example,[4] developed an ambient facial interface used toprovide visual feedback and confirmation to theuser in a manner independent of age, language,culture and mental alertness. Their work usedanimated or actual human faces to display emo-tional expressions easily recognizable by anelderly person with non-verbal feedback. Thefacial responses were controlled by measure-ments or interpreted data on the user’s currentstate or the products and objects with which heor she is interacting. By combining these mea-surements into a single facial expression, whichis displayed to the user it assisted the user toevaluate their task. Reference [5] commentatedon the growing area of application development.Their system moved application creation fromthe developer’s perspective whose focus was ondevices and their interactions, to end users’interpretation of goals or tasks. Comic scriptsscenarios depicting assistive aids were presentedto participants to explain the applications intheir own words. This study highlighted the sep-aration between the ages and living conditions ofthe designers and the people using AAL ser-vices. To overcome the disparity end user coulddevelop applications using a text-based interface.End users would drag and drop scenarios of

3 http://www.quietcaresys-tems.co.uk

Figure 2. Illustrating sensor activations per day split into time interval [6].

Week

Sensor Firings

07

200

0 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 06 05

Sleeping

Zone

Early morning Late morning Lunch Afternoon Evening Late evening

400

600

800

1000

1200

1400

1600

1800

2000 First medication

change Second medication

change

The AAL servicesstore personal

activity profiles overperiods of time.

The trend of userbehavior and thepattern of user activity events provide rich

information in theanalysis of care

recipients’ behaviorpatterns.

MULVENNA LAYOUT 12/3/10 8:34 PM Page 5

IEEE Communications Magazine • January 20116

daily living using a list of predefined words intoa window and click a run button. The systemwould then parse the words and develop a speci-fication for an application to meet the needs ofclients involved.

Reference [6] modeled ADLs for care recipi-ents, and found that using these life patternscould provide enhanced home-based care (Fig.2). The study found that changes in a busynessmetric were visible and detectable even withirregular behavior.

Reference [7] indicated that a graphical userinterface significantly improved performanceresponse times, error rates, and satisfaction rat-ings compared with an existing text based inter-face for nurses using an information system. Thestudy highlighted that staff found the advent of amore functional, intuitive and user friendly GUIeasier to learn to aid navigation through the sys-tem and to prioritise management tasks.

CASE STUDY: NOCTURNAL AAL SERVICES FOR

PEOPLE WITH DEMENTIA

The University of Ulster and the Fold HousingGroup collaborate in the NOCTURNAL projecton issues including data visualization for AALservices. The goal is to develop a solution thatsupports older people with mild dementia intheir homes, specifically during the hours ofdarkness. This is a relatively new area of researchand was identified as a key area of need for carerecipients with dementia. It is also of interestbecause of the negative impact that lack of sleepand consequent anxiety causes for the informalcarer in the home of the person with dementia.In their literature review on night-time care ofpeople with dementia using AAL services, [8]found in that only 7 percent of papers addressed

Table 2. How visualization supports decision support across different users and roles.

Data

Visualization!

Decision support

User typeData capture/viewing/analysis

Dataabstraction

Communicationof advice

Providinginformation

Managingnuances of alarmescalation

Care recipient

Data capturedfrom sensorsusing internalinfrastructurenetwork; userresponses

Graphical(simple,metaphor for day,e.g., traffic lights)

Advice onactivities ofdaily living

Self care

Supply instruction;multimodal;communication(e.g., audible)

Formal carerson-site

Real-time,archive

Abstracted textinformation onsingle carerecipients

Graphical(summary daychart +weekly/monthlytrend)

Support forassessments

Assessmentinferenceson carerecipient

Formal carersoff-site (contactby networks)

Real-time,archive

Abstracted textinformation ongroups of carerecipients

Graphical(summary daychart +weekly/monthlytrend)

Assessmentinferenceson carerecipient(s)

Contact by mobilenetworks or broad-band networks

Telecare/Tele-health monitorsoffsite (contactby networks)

Real-time,archive

Graphical(summary daychart +weekly/monthlytrend + statisticalmetrics)

Abstracted alarmdashboards forsingle and groupsof care recipients;Contact by mobilenetworks

Informal carerson-site

Non-medicalinformationabstracted forinformal carers

Graphical(simple,metaphor for daye.g., traffic lights)

Informal carersoff-site (contactby networks)

Non-medicalinformationabstracted forinformal carers

Graphical(simple,metaphor for daye.g., traffic lights)

Contact by mobilenetworks

Technical mainte-nance (networkaccess for remotemonitoring)

Anonymousaccess to data;time restricted;real-time analysis

Status reports

MULVENNA LAYOUT 12/3/10 8:34 PM Page 6

IEEE Communications Magazine • January 2011 7

night-time specific issues with a further 26 per-cent focusing on night and day activities togeth-er. Of the night-time specific papers, only halfinvolved any form of visual data representation,indicating that there is need for research in thisarea.

The focus on night-time AAL services centersaround lighting and guidance, motion monitor-ing, and intervention decision-support. In thecase study, the intention is for the AAL servicesto provide reassurance, aid and guidance for thegeneral behavior of the care recipient and tosupport a stable circadian rhythm. This casestudy is appropriate to illustrate the issues incommunication between AAL services and thedifferent user types and roles — as identified inTable 1 — who all are potential actors. It alsoprovides a clear example of the need for con-verged communication networks, as it uses abroad range of sensors, processors and actuators.In this case, the following communication net-works and protocols are used:• Body sensor and personal area networks,

relying on protocols such as Bluetooth, Zig-bee, and Ant+

• Home networks, providing either wired andwireless connectivity with specialist proto-cols such as x10 used to supplement stan-dard 802.x networks

• Wider connectivity to remote stakeholders(cares, monitors and healthcare profession-als) via secure broadband services

• Enhanced cellular services (GPRS, 3G,UMTS, Long Term Evolution and beyond)facilitating a sort of virtual presence for fam-ily members and occasional carers.The visual representation of the data is a key

component of the work, and visualization isdesigned based on the needs of the different

users and their roles. Table 2 shows the resultsof an analysis of the users and their roles fromthe perspective of the translation of data to amode where it supports decision-making andhow visualization plays a part in that process.

The visualization of AAL data is accessibleon computer interfaces after authenticatedaccess, but as Table 2 shows, different modalitiesof access are also possible. For example, alertscan be pushed to mobile clients as the alerts aretriggered. Table 2 illustrates how the AAL ser-vices generate data, which is then abstractedbefore being made available for visualization insupport of decision support. The three maindecision support modes are: communication ofadvice, provision of information, and manage-ment of (nuances of) alarm escalation. Figure 3shows the Tunstall-based interface for a techni-cal maintenance user at single sensor level aswell as multi-sensor over a 28-day period.

The need to visualize information temporallyis clear in each illustration in Fig. 3 as is therequirement to show activities in different loca-tions of the home of the care recipient. This typeof visualization of activities across space andtime displayed on two dimensions of the inter-face, as used in our project, is becoming themost common manner of displaying the wellbe-ing of the care recipient in many AAL services.However, this interface is not appropriate forthe care recipient or carers using AAL services.

A metaphor interface, for example, asdescribed by [9], would be more appropriate tothe abstraction and appreciation of the on-goingAAL supported care. In their research, the inter-face (Fig. 4) was in a mobile device to monitorwellbeing, and it incorporate images such asflowers and butterflies that change as the personbecame more active in daily exercise regimes.

Figure 3. Chart illustration of sensor data: a single sensor cell (left), location vs. time display and previous 28 days cell (right).

MULVENNA LAYOUT 12/3/10 8:34 PM Page 7

IEEE Communications Magazine • January 20118

DISCUSSIONIn developed countries such as the UK, theincreasing prevalence of chronic disease in theageing population provides the context for morepervasive deployment of AAL services. Addi-tionally, government policy such as the citizen’sright to high-speed connectivity by 2020 through-out the United Kingdom as published in [10]provides the social and technological climate forsuch applications to succeed. However, there is asignificant risk that due to the large streams ofdata generated by these AAL services, poor visu-alization techniques in interfaces, as described,will become a key issue, leading to potential mis-understandings and misinterpretation of infor-mation and data generated from AAL services.

This article has described the systems thatprovide AAL services in support of assisted liv-ing. It has examined exemplars in both commer-cial and academic areas. The NOCTURNALresearch project was used to highlight the visual-ization components required to service the needsof the different user types. The paper highlightsthat a key requirement for successful AAL ser-vice uptake is to address the needs of the differ-ent user groups and translate these needs intointerface and data visualization components thatcommunicate clearly the wellbeing and health ofthe care recipient(s).

It is evident that there is currently a gap withregards to the supply of a fully functional,dependable and appropriate visualization of rel-evant service data, particularly with regards tothe different user types and roles of peopleinvolved in the delivery of care. There is a needto design applications, which display AAL ser-vice data in a meaningful, holistic yet concisemanner. Identifying the needs for each usergroup and how these needs are provided bothphysically and visually has to be resolved if AALservices are going to be utilized more fully insociety. Gil et al. [6] suggested that the focus ofvisual representation should concentrate on liv-

ing aspects that are regular, e.g., sleeping, eatingetc., and have a relationship with wellbeing. Akey requirement for AAL services is to minimizethe cognitive overhead required to interpretinformation, by presenting normal behaviors andpatterns in the most succinct form possible, andhighlighting abnormal behaviors and alarmstates tailored to each end user group’s needs.The work by Consolvo et al . [9] on visualmetaphors that requires minimal cognitive pro-cessing by the users is promising and indicatespossible development pathways that communi-cate regular behaviors and useful feedback tocare recipients and their informal carers.

ACKNOWLEDGMENTWe would like to acknowledge the support ofthe carers and people with dementia in NorthernIreland who contribute to the work of the Noc-turnal project, which is supported by the UnitedKingdom Research Councils and the TechnologyStrategy Board’s Assisted Living InnovationPlatform under grant award TS/G002452/1.

REFERENCES[1] K. Ducatel et al. (Eds.), “Scenarios for Ambient Intelli-

gence in 2010,” IPTS-ISTAG, EC: Luxembourg, 2001;www.cordis.lu/ist/istag.

[2] D. J. Cook, J. C. Augusto, and V. R. Jakkula, “AmbientIntelligence: Technologies, Applications, and Opportuni-ties,” Pervasive Mobile Comp., vol. 5, no. 4, 2009, pp.277–98.

[3] M. Wojciechowski and J. Xiong, “A User Interface LevelContext Model for Ambient Assisted Living,” SmartHomes and Health Telematics, 2008, pp. 105–12.

[4] B. Takacs and D. Hanak, “A Mobile System for AssistedLiving with Ambient Facial Interfaces,” Int’l. J. Comp.Sci. Info. Sys., vol. 2, no. 2, 2007, pp. 33–50.

[5] K. Truong, K., E. Huang, and G. Abowd, “CAMP: AMagnetic Poetry Interface for End-User Programming ofCapture Applications for the Home,” Proc. Int’l. Conf.Ubiquitous Comp., Nottingham, UK, 2004, pp. 143–60.

[6] N. M. Gil et al., “Data Visualization and Data MiningTechnology for Supporting Care for Older People,”Proc. 9th ACM SIGACCESS Int’l. Conf. Comp. Accessibil-ity, Tempe, AZ, 2007, pp. 139–46.

[7] N. Staggers and D. Kobus, “Comparing Response Time,Errors, and Satisfaction between Text-Based and Graph-ical User Interfaces During Nursing Order Tasks,” J.American Medical Info. Assoc ., vol. 7, 2000, pp.164–76.

[8] W. Carswell et al., “A Review of the Role of AssistiveTechnology for People with Dementia in the Hours ofDarkness,” Tech. Healthcare, vol. 17, no. 4, 2009.

[9] S. Consolvo et al., “Flowers or a Robot Army?: Encour-aging Awareness & Activity with Personal, Mobile Dis-plays,” Proc. 10th Int’l. Conf. Ubiquitous Comp., Sept.21–24, 2008, Seoul, Korea.

[10] Digital Britain, “Final Report, Department for Culture,Media and Sport and Department for Business, Innova-tion, and Skills,” TSO, June 2009.

BIOGRAPHIESMAURICE MULVENNA [SM] (md.mulvenna@ulster.ac.uk) is aprofessor of computer science at the TRAIL living laborato-ry in the University of Ulster. He publishes in the area ofpervasive computing in support of older and disabled peo-ple, and reviews for IEEE Pervasive Computing and IEEEPervasive Computing and Applications.

WILLIAM CARSWELL is a researcher at the University of Ulster.He has experience in developing intelligent systems to col-lect and analyse sensory data and provide appropriateautomated responses to cater for detected abnormal pat-terns. He is an IEEE affiliate member.

PAUL MCCULLAGH is a reader in computing and mathematicsat the University of Ulster. He is on the Board of the Euro-pean Society for Engineering and Medicine and BCS Health

Figure 4. Garden mappings and two sample gardens [9].

MULVENNA LAYOUT 12/3/10 8:34 PM Page 8

IEEE Communications Magazine • January 2011 9

Northern Ireland. His interests include health informaticseducation, open health, interface design, and assisted liv-ing applications. Current research projects include EU FP7BRAIN: BCIs with Rapid Automated Interfaces for Nonex-perts, where he is ethics manager; and EPSRC SMART2 —Self Management Supported by Assistive, Rehabilitation,and Telecare Technologies.

JUAN CARLOS AUGUSTO [M] focuses on the design and imple-mentation of intelligent environments, to which he hascontributed more than 100 publications, including theHandbook on Ambient Intelligence and Smart Environ-ments. He is Editor-in-Chief of the Book Series on AmbientIntelligence and Smart Environments, Program Chair of Per-vasive Healthcare 2011 and IE ‘11, and Co-Editor-in-Chiefof the Journal on Ambient Intelligence and Smart Environ-ments.

HUIRU ZHENG is a lecturer in computer science at the Univer-sity of Ulster. She is an active researcher in the broad areaof biomedical informatics and has published over 90 scien-tific papers. Her main research interests include data min-ing, machine learning decision support, patternrecognition, intelligent data analysis, gait analysis, andactivity monitoring.

PAUL JEFFERS works for Fold Housing Association as a techni-cal researcher. He has a special interest in developing solu-tions to help deal with the ever-increasing demands fromproviding services to people with dementia. He has experi-ence in sensor and vision systems, and is currently workingon the NOCTURNAL and VirtEx ALIP projects.

HAIYING WANG is currently a lecturer in the School of Com-puting and Mathematics at the University of Ulster. Hisresearch focuses on artificial intelligence, machine learning,pattern discovery, and visualization, and their applicationsin medical informatics and bioinformatics.

SUZANNE MARTIN is an occupational therapist and, as a read-er, is a full-time member of academic staff at the Universi-ty of Ulster. Her research investigates the use of new andemerging technologies in health and social care. She isinterested primarily in research methods that promote end-user participation to explore complex interventions. She isalso a contributor to the Cochrane Library synthesizing theevidence base for a range of healthcare interventions.

MULVENNA LAYOUT 12/3/10 8:34 PM Page 9