Responsible Development of Self-learning Assisted Living Technology

for Older Adults with Mild Cognitive Impairment or Dementia

Evi Zouganeli

, Flávia Dias Casagrande

, Torhild Holthe

, Anne Lund

, Liv Halvorsrud

Dag Karterud

, Adele Flakke-Johannessen

, Hilde Lovett

, Sindre Kjeang Mørk

Jørgen Strøm-Gundersen

, Erik Thorstensen

, Reidun Norvoll

, Ruud ter Meulen

Mari-Rose Kennedy

Richard J. Owen

, Miltos Ladikas

and Ellen-Marie Forsberg

Oslo & Akershus University College of Applied Sciences, Dept. Electronics and IT, Pilestredet 35, 0166 Oslo, Norway

Oslo & Akershus University College of Applied Sciences, Dept. Occupational Therapy, Pilestredet 48, 0167 Oslo, Norway

Oslo & Akershus University College of Applied Sciences, Dept. Nursing, Pilestredet 32, 0166 Oslo, Norway

The Norwegian Board of Technology, Kongens gate 14, 0153 Oslo, Norway

Sensio AS, Brugata 19, 0186 Oslo, Norway

Oslo & Akershus University College of Applied Sciences, Work Research Institute, Stensberggata 25, 0170 Oslo, Norway

University of Bristol, Whatley Road 39, Bristol, U.K.

Exeter Business School, Rennes Drive, Exeter, U.K.

Karlsruhe Institute of Technology, P.O. Box 3640, Karlsruhe, Germany

Keywords: Self-learning Systems, Assisted Living Technology, Welfare Technology, Smart Home, Older Adults, User

Involvement, Independent Living, Responsible Research and Innovation (RRI).

Abstract: In this paper we present work in progress in the Assisted Living Project – responsible innovations for dignified

lives at home for people with mild cognitive impairment or dementia. The project has a distinctly

interdisciplinary approach and engages experts in nursing and occupational therapy, in ethics and responsible

research and innovation, and in technology, in particular automation and machine learning. Our approach is

to involve the end-users, their family and their care providers and develop technology responsibly together

with them. The technological approach employs self-learning systems to develop solutions that provide

individualised support in accordance with the user’s values, choices, and preferences. The paper presents our

approach, current findings and future plans.

1 INTRODUCTION

We report work in progress from the Assisted Living

project, an interdisciplinary project that aims to

develop technological solutions to support older

adults with mild cognitive impairment or dementia

(MCI/D) live a safe and fulfilling life at home, with

dignity and independence. The project engages

experts in nursing and occupational therapy, in ethics

and responsible research and innovation (RRI), and in

technology, in particular automation and machine

learning (ML). Solutions will be developed together

with the users and tried out in field trials with around

10-15 end-users.

The main incentive for assistive living technology

(ALT) solutions was originally to enable those in

need of medical care remain at home and hence

reduce costly stays in staffed care units. This can be

the case for both outpatients recuperating after for

example an operation or an accident, people with

chronical diseases, as well as the elderly and people

with special needs (Aspnes et al., 2012). Today there

is a plethora of home automation and wearable

technology to support everyday activity needs related

to security, safety, communication, work, social

contact, exercising, entertainment, and other. The

strong synergies between the care-related and the

mainstream home and lifestyle automation, with an

enormous growth potential, have led to a

204

Zouganeli, E., Casagrande, F., Holthe, T., Lund, A., Halvorsrud, L., Karterud, D., Flakke-Johannessen, A., Lovett, H., Mørk, S., Strøm-Gundersen, J., Thorstensen, E., Norvoll, R., Meulen,

R., Kennedy, M-R., Owen, R., Ladikas, M. and Forsberg, E-M.

Responsible Development of Self-learning Assisted Living Technology for Older Adults with Mild Cognitive Impairment or Dementia.

DOI: 10.5220/0006367702040209

In Proceedings of the 3rd International Conference on Information and Communication Technologies for Ageing Well and e-Health (ICT4AWE 2017), pages 204-209

ISBN: 978-989-758-251-6

consolidation of solution providers across these two

segments and a revisited interest in ALT in recent

years.

Commercial devices have attained a certain

degree of maturity. There is also a formidable and

steadily increasing list of “stand alone” applications.

The number of integrated systems that provide

seamless and holistic solutions both in the home and

ubiquitously is, however, relatively small and the

degree of integration is still quite limited.

Standardisation and regulation are now promoting an

open model where applications and devices from an

ecosystem of providers shall be possible to plug in on

demand. An open model will facilitate a flexible

environment where some of the devices and services

are provided by the national health system or a health

insurance, and others are purchased by the individual.

This is, however, far from the case at the moment as

solutions are more of a non-interoperable proprietary

patchwork.

This position paper presents our approach to

developing solutions for people with MCI/D as well

as some of our current findings. Our approach builds

upon two main hypotheses/ positions:

i. When developing ALT solutions it is vital to

involve all user groups and all stakeholders –

throughout the process and right from the start. In our

opinion this is especially important in the case of

people with MCI/D and other groups with cognitive

impairments and similarly other vulnerable groups

whose ability to contribute is underestimated and

hence their voice tends to be overheard.

ii. The use of self-learning systems can provide

people with cognitive impairments with the

appropriate degree of cognitive enhancement and

enable them continue to live independently, in

accordance with their values, personal choices, and

individual needs. Indeed, each person is an individual

and “one-size-fits-all” types of solutions are by

definition quite unlikely to serve the individual well

and in all circumstances.

In the following we present some more details

regarding our approach as well as current evidence

that supports these positions.

1.1 Background

Politicians and health care providers share today great

optimism regarding the potential of emerging

technology to support older people at home.

Technology is expected to reduce the pressure on

needs for public health services, and to contribute to

independence and dignity for older people with mild

cognitive impairment and early phase of dementia

(Lindqvist et al., 2013; Nygård and Starkhammar,

2007; Øderud et al., 2015). However, matching

technology to a person’s needs successfully, depends

upon several things: the ability to reveal needs for

support in the “subject of care”; the degree of

individualization to the user’s needs and context; the

maturity and user-friendliness of the technology; and

the robustness and predictability of the technology as

sustainable solutions (Arntzen et al., 2016; Jentoftet

al., 2014; Winblad et al., 2004). Further, an important

factor concerns creating a supportive network for the

user (Rosenberg et al., 2012). Therefore,

investigating the potential of current technologies to

support older adults, and in particular how to

individualize such devices/solutions to address

individual needs and preferences, is a vital

component for developing useful new services.

Our approach is to involve the residents in a

seniors’ care dwelling, by discussing their habits,

needs and preferences, as well as their experience

with current technology, in order to identify possible

pitfalls and success criteria.

1.1.1 Mild Cognitive Impairment

Participants in our study may have Mild Cognitive

Impairment (MCI) or be in an early phase of

dementia. Cognition encompasses attention,

concentration, memory, comprehension, reasoning,

and problem solving. Mild cognitive impairment was

described by Winblad et al. (2004) to be a useful term

as both a clinical and research entity. MCI is more

than a pre-clinical stage of Alzheimer’s disease. MCI

may 1) progress over time 2) be stable, or 3) the

person may recover. Risks of mortality seems high for

all types. Hedman et al. (2013) studied patterns of

functioning in older adults with MCI and found that

they exhibited different patterns; stable, fluctuating,

descending or ascending patterns. The patterns may

change over time, and thus individual support is

needed (Hedman et al., 2013).

1.2 User and Stakeholder Involvement

User involvement can be conducted for epistemic,

normative and/ or instrumental reasons. (Fiorino,

1990). Our project epistemically aims at “co-

production of knowledge”. This is defined as being

engaged in the process of mutual learning, and taking

part in identifying solutions (Askheim, 2016). This is

in line with the normative idea of empowerment. The

participants are given the authority to decide what is

right for them. This indicates that power relations are

changed, the person is actively involved, and

Responsible Development of Self-learning Assisted Living Technology for Older Adults with Mild Cognitive Impairment or Dementia

205

perceived as an expert on own health and life (Tveiten

and Knutsen, 2011).

The user-centered approach is also embedded in

the Responsible Research and Innovation (RRI)

methodology (Forsberg et al., 2015; von Schomberg,

2013). The most central values are according to Owen

et al. (2013) reflection on the intersections between

science and society; clear distribution of

responsibility for future events, built-in precautionary

measures; and discussions over the intent of research

and innovation. The RRI framework applied in this

project has four integrated dimensions: Anticipation,

Reflexivity, Inclusion and Responsiveness (Stilgoe et

al., 2013). Central to RRI is an idea of mutual

transdisciplinary learning and taking part in

identifying solutions (Wickson and Carew, 2014).

Porcari et al. (2015) distinguish between designing

for users and designing with users, where

participatory development with users is a

“responsible approach” finally leading to more

acceptable products.

2 USER INVOLVEMENT

We employ a combination of techniques and methods

in order to understand the users’ preferences and

needs related to challenges in everyday living, the use

of and attitudes towards technology, and their

perceptions of own health. We combine the use of

standardized questionnaires (such as Rand-36,

Hospital Anxiety Depression Scale, Lawton and

Brody ADL), dialogue cafés and semi-structured

individual interviews. Also focus groups with staff

are performed.

We use a stepwise process in order to invite,

recruit and retain participants in the study. The

process aims to engage the residents in the seniors’

care dwelling and involve them as much as possible

during the intervention.

Introductory discussions with the leaders and

housekeepers in the seniors’ home were used to

anchor the project. All residents were then invited to

a presentation of the project during one of the regular

“house meetings”. Approximately 20 residents

consented and participated in semi-structured

interviews with questionnaires about technology,

perception of health, memory and quality of life. The

researchers showed up at the seniors’ care dwelling

approximately 2-3 times a week in the beginning, to

get acquainted with the seniors, and inform about the

project.

All residents were invited to a first dialogue café

(DC1) to discuss challenges they experienced in their

daily life. The dialogue café method was developed

with inspiration from several methods for user

involvement; scenario workshops (Barland, 2013),

dialogue conferences (Pålshaugen, 1998) and world

café. In addition to obtaining information on needs,

the dialogue café method stimulates for peer learning.

DC1 was organized as group discussions, using user

stories to help the residents relate. We wanted the

DC1 to be as open as possible without a technology

aspect. At a second dialogue café (DC2) we discussed

examples of technological solutions. We designed

user stories with cartoons to facilitate the group

discussions with the residents. The choice of user

stories and the group discussions at DC2 reflected

both technical alternatives and ethical considerations.

For example, are the residents willing to allow a

camera in their home; who can have access to the

images; are the residents willing to be localized in

order to facilitate social contact, and under what

conditions.

Dialogue cafés have so far proved to be an

efficient and creative way for engaging the seniors,

presenting ideas and thoughts, stimulate peer

learning, and for understanding and discussing both

challenges and solutions. In particular, it has been

very useful for the project team to receive immediate

feedback on whether the suggested user stories were

of any interest for the residents at all. This directed

the work onwards.

We will further invite the residents to new

dialogue cafés for presenting and discussing concrete

solutions, and to proceed in a similar manner to

ensure that the residents are involved throughout the

development process.

Further discussions and individual interviews

with the residents are planned to reveal the individual

needs and wishes beyond what is expressed in a

public and social setting. In addition, further

discussions regarding the detailed features of

different technical alternatives are required in order to

identify solutions for the first trial. These are required

to meet needs, abide to wishes and choices, as well as

be within the project’s resource and technical

constraints.

3 SUMMARY OF FINDINGS

3.1 User Needs

The most prominent user needs that resulted from

DC1 related to eight areas:

ICT4AWE 2017 - 3rd International Conference on Information and Communication Technologies for Ageing Well and e-Health

206

1. Falls – the fear of falling, injury and not getting

help. Some of the residents have a security/pendent

alarm button that is provided by the national health

system. Although help shows up within short time, a

number of limitations – only operating indoors,

requiring consciousness, not knowing whether the

alarm actually has been received and when help is

coming – are major shortcomings.

2. Being outdoors and access to fresh air. Being safe

when out of the house was crucial. Also physical

mobility and secure and predictable transportation are

important.

3. Ability to orient oneself at night. Dark

environment and possibly impaired vision may

influence navigation/orientation at night, e.g. for a

toilet visit, and can increase the risk of falling.

4. “Button-phobia”. Technology can be difficult to

use, due to small buttons and unfamiliar interfaces, as

well as passwords and codes.

5. Social contact with others, both inside and outside

the seniors’ care dwelling can be challenging.

6. Safety at home. This is multifold and associates to

not always getting help when required and within

short time, access to their apartment by helpers even

if the door is locked.

7. Sleeping sufficiently and well. Challenges include

the difficulty to fall sleep, waking up frequently

during the night, and/ or waking up too early in the

morning.

8. Self-sufficiency and autonomy. Even if the

residents do feel relatively autonomous and self-

sufficient, their daily routines need to conform to the

schedule of others. The schedule of family, nurses

and staff can compromise the individual’s preferred

activities and daily routines, introduce long waiting

and create unpredictability and diminished control

over own life.

3.2 Priorities

A summary of the main findings will be presented at

the conference whereas the details of these will be

described in a separate publication. Some of the key

reflected characteristics were:

i. The high importance/ priority of being independent,

self-sufficient, and in control over own life. Most of

the residents were also wary of troubling their family

and friends.

ii. The wish to remain active and a fear that relying

on help from others or from a system may cause a

deterioration of their cognitive ability.

iii. A willingness to trade-off privacy for better safety.

3.3 Current Experience with User

Involvement

Our overall experience with user involvement so far

supports our original hopes and expectations. Indeed,

both the care providers and the residents themselves

have provided us with invaluable feedback. Note that

next-of-kin have not yet been interviewed. In several

occasions the research team was reminded of how

difficult it is to understand the needs and preferences

of other people and speak for others. This is the case

despite the best intentions, a lot of expertise, and even

personal experience through the researchers’ own

elderly family members. Indeed, many of our

predictions regarding which solutions would appeal

to the residents were quite wrong.

4 THE ROLE OF SELF-

LEARNING SYSTEMS

Typically, commercial smart-home solutions provide

assistive devices such as reminders, calendar, night

lights, electric cooker timers, medicine dispenser,

picture phone, etc. (Topo et al., 2004; Jones, 2004).

More complete solutions integrating several

functionalities have also been developed. The

portable device in the COGKNOW project provided

memory-aids, social contact (e.g. picture dialing),

help on daily activities (e.g. lamp control), and safety

functions (Mulvenna et al., 2010). This was

integrated with two additional systems for the Rosetta

project (Hattink et al., 2014). This system recorded

behaviour patterns to analyse sleep-awake rhythms,

mobility, meal preparation and personal hygiene. It

also detected emergency situations such as falls and

alarmed carers. In general, these systems were well

received, especially when introducing functionality

that enhanced the feeling of safety.

Especially in the past five-ten years there has been

an emergence of solutions that employ machine

learning (ML). ML is used for example for better fall

detection (Choi et al., 2011), automatic activity

recognition (Chen et al., 2010), or to monitor/ study

behaviour patterns (Cook et al., 2015). ML has been

also used to generate prompts to assist daily activities

in the CASAS smart home (Das et al., 2012). A

number of projects address the difficulty of executing

daily activities. For example, the COACH system

(Hoey et al., 2007) assists people with dementia in the

hand-washing activity via a camera and provides

automatic cues to assist activity completion. Feki et

al. (2009) deploy several sensors to monitor activity

Responsible Development of Self-learning Assisted Living Technology for Older Adults with Mild Cognitive Impairment or Dementia

207

execution (i.e. meal preparation and eating) and issue

automatic prompts in case of error. Karakostas et al.

(2015) in the Dem@Care project developed a

semantically integrated multi-sensor system that

provides holistic support and tested it on one

dementia patient.

Yet the potential of ML is largely untapped. The

self-learning and self-adapting potential of ML-

techniques are important characteristics that enable

individualised solutions without the need to manually

tailor individually – a process that is prohibitively

costly in traditional systems. With ML the system can

in principle observe, learn and adapt accordingly on

its own. As presented in section 1.1, MCI/D is an

example of a condition that is more of an individual

spectrum of characteristics and impairments rather

than one simple to define condition. Here every

person is indeed an individual case and pre-fitting the

system to the user is not only costly but in reality quite

limited, if at all possible. Moreover, the condition

progresses in an individual manner and at an

individual speed. This demands a technology that can

sense and adapt to evolving needs. In addition, a self-

learning system is potentially capable to evolve and

meet a set of preferences and requirements that are

latent and to an extent unknown to both the user, and

the care provider/ health expert.

Beyond the personalisation of services and care,

there is an untapped potential for higher level

semantic system intelligence and cognitive

enhancement on the person’s own terms. The idea is

that a system shall comprehend the overall situation –

the objective facts, potential hazards, as well as the

person’s subjective experience and personal choice –

and support the human achieve their current goal.

The idea of the smart-home has been gradually

evolving from that of a simple automation towards

the vision of an unobtrusive interconnected

environment, that is sensitive and adaptive to the

inhabitants’ needs and behaviour (Aarts and Wichert,

2009). This extends both within and outside the home

following new possibilities in the advent of the

Internet of Things (Zouganeli and Svinnset, 2009).

Current smart-home paradigms rely on creating good

user-interfaces and voice-interfaces are positive steps

to that end.

A new paradigm can envisage systems that

understand the intention of the human, anticipate

hurdles, device solutions, predict outcomes, and are

an extension of the human on the human’s terms. This

creates of course new challenges as autonomy, safety,

privacy and ethical considerations need to be

thoroughly safeguarded throughout. The rapid

advance of artificial intelligence and adjacent fields

may enable such solutions in the not too far future.

5 SUMMARY

We have presented work in progress and our

approach to developing technological solutions for

people with MCI/D. We argue that the users need to

be involved all the way in order to develop good

systems, and this holds not least for people with

MCI/D. Our preliminary experience supports this

view. We have also made the case for self-learning

systems as well as presented our vision regarding the

future evolution of these.

ACKNOWLEDGEMENTS

The authors wish to thank the residents and the

housekeepers in the seniors’ care dwelling at Skøyen

Omsorg+ and in Oslo Municipality. The project is

financed by the Norwegian Research Council under

the SAMANSVAR programme (247620/O70).

REFERENCES

Aarts, E. & Wichert, R., 2009. Ambient intelligence: A

Survey. Technology Guide, 43(4), pp.244–249.

Arntzen, C., Holthe, T., & Jentoft, R. 2014. Tracing the

successful incorporation of assistive technology into

everyday life for younger people with dementia and

family carers. Dementia: The International Journal of

Social Research and Practice, 15(4).

Askheim, O. 2016. Samproduksjon som velferdssektorens

kinderegg. Tidsskrift for omsorgsforskning, 2(1), 24-36.

Aspnes, E. et al. 2012. Velferdsteknologi i boliger,

Available: http://www.sintef.net/globalassets/project/

velferdsteknologi/velferdsteknologi_i_boliger_rapport

_revidert_120312.pdf [Accessed January 4, 2017].

Barland, M. 2013. Scenario workshop method description,

Available: http://www.pacitaproject.eu/wp-content/

uploads/2015/01/PACITA-D-6-1-Method-description

_stakeholder-scenario-workshop_v4-_2_.pdf

Chen, C., Das, B., Cook, D.J, 2010. A data mining

framework for activity recognition in smart

environments. Procs. of the Sixth International

Conference on Intelligent Environments, pp 80-83.

Choi, Y., Ralhan, A.S. & Ko, S., 2011. A study on machine

learning algorithms for fall detection and movement

classification. 2011 International Conference on

Information Science and Applications, ICISA 2011,

(May 2011).

Cook, D.J., Schmitter-Edgecombe, M., Dawadi, P., 2015.

Analyzing activity behavior and movement in a

ICT4AWE 2017 - 3rd International Conference on Information and Communication Technologies for Ageing Well and e-Health

208

naturalistic environment using smart home techniques.

IEEE J Biomed. Health Informatics, 19(6), pp 1882-92.

Das, B. et al., 2012. PUCK: An automated prompting

system for smart environments: Toward achieving

automated prompting-challenges involved. Personal

and Ubiquitous Computing, 16(7), pp.859–873.

Feki, M.A., Biswas, J. & Tolstikov, A., 2009. Model and

algorithmic framework for detection and correction of

cognitive errors. Technology and Health Care, 17(3),

pp.203–219.

Fiorino, D. J. 1990. Citizen Participation and

Environmental Risk: A Survey of Institutional

Mechanisms. Science, Technology, & Human Values,

15(2), 226–243.

Forsberg, E., Quaglio, G., O'Kane, H., Karapiperis, T., Van

Woensel, L., & Arnaldi, S. 2015. Assessment of science

and technologies: Advising for and with responsibility.

Technology in Society, 42, 21-27.

Hattink, B.J.J. et al., 2014. The electronic, personalizable

Rosetta system for dementia care: exploring the user-

friendliness, usefulness and impact. Disability and

rehabilitation. Assist. Technology, 3107(Mci), pp.1–11.

Hedman, A., Nygard, L., Almkvist, O., & Kottorp, A. 2015.

Patterns of functioning in older adults with mild

cognitive impairment: A two-year study focusing on

everyday technology use: Corrigendum. Aging &

Mental Health, 19(6), 569.

Hoey, J. et al., 2010. Automated handwashing assistance

for persons with dementia using video and a partially

observable Markov decision process. Computer Vision

and Image Understand., 114(5), pp.503–519.

Jentoft, R., Holthe, T., & Arntzen, C. 2014. The use of

assistive technology in the everyday lives of young

people living with dementia and their caregivers. Can a

simple remote control make a difference? Int

Psychogeriatr, 26(12).

Jones, K., 2004. Enabling Technologies for People with

Dementia. Report of the assessment study in England.

Karakostas A., Lazarou I., Mediskos G., Stavropoulos T.

G., Kompastianis I., Tsolaki M. 2015. Sensor-based in-

home monitoring of people with dementia using remote

web technologies. Int. Conf. on Interactive Mobile

Comm. Tech. and Learning (IMCL). Thessaloniki, 19-

20 Nov. 2015.

Mulvenna, M. et al., 2010. Designing & Evaluating a

Cognitive Prosthetic for People with Mild Dementia.

Proceedings of the 28th Annual European Conference

on Cognitive Ergonomics - ECCE ’10, (August), p.11.

Lindqvist, E., Nygard, L., & Borell, L. 2013. Significant

junctures on the way towards becoming a user of

assistive technology in Alzheimer's disease. Scand J

Occupational Therapy, 20(5).

Nygård, L., & Starkhammar, S. 2007. The use of everyday

technology by people with dementia living alone. Aging

& Mental Health, 11.

Owen, R., Stilgoe, J., Macnaghten, P., Gorman, M., Fisher,

E., & Guston, D. 2013. A Framework for Responsible

Innovation. In R. Owen, J. Bessant, & M. Heintz (Eds.),

Responsible Innovation (pp. 27–50). John Wiley &

Sons, Ltd.

Porcari, A., Borella, E., & Elvio, M. 2015. A Framework

for implementing Responsible Research and

Innovation. In ICT for an ageing society (No. D2.4).

Italian Association for Industrial Research (AIRI)

/Responsible-Industry.

Pålshaugen, Øyvind. 1998. Organization development

through development organization. In: Development

coalitions in working life, ed. Bjørn Gustavsen, Tom

Colbjørnsen, and Øyvind Pålshaugen. Amsterdam:

John Benjamins Publishing Company.

Rosenberg, L., Kottorp, A., & Nygard, L. 2012. Readiness

for technology use with people with dementia: The

perspectives of significant others. J Appl. Gerontology,

31(4).

Stilgoe, J., Owen, R., & Macnaghten, P. 2013. Developing

a framework for Responsible Innovation. Research

Policy 42.

Topo, P., Saarikalle, K., Holthe, T., et al., 2004. Enabling

Technologies for People with Dementia. Report of

Picture Gramophone assessment: National findings

from Finland, Ireland, Norway and UK and cross

national results.

Tveiten, S., & Knutsen, I. 2011. Empowering dialogues -

the patient's perspective. Scand J of Caring Sci, 25.

Von Schomberg, R. 2013. A Vision of Responsible

Research and Innovation. In R. Owen, J. Bessant, & M.

Heintz (Eds.), Responsible Innovation (pp. 51–74).

John Wiley & Sons, Ltd.

Wickson, F., & Carew, A. L. 2014. Quality criteria and

indicators for responsible research and innovation:

learning from transdisciplinarity. Journal of

Responsible Innovation, 1(3), 254–273.

Winblad, B., Palmer, K., Kivipelto, M., Jelic, V.,

Fratiglioni, L., Wahlund, L., Petersen, R. 2004. Mild

Cognitive Impairment - beyond controversies, towards

a consensus: report of the International Working Group

on Mild Cognitive Impairment. J of Internal Medicine,

256.

Zouganeli E., Svinnset I. E., 2009. Connected Objects and

the Internet of Things – a Paradigm Shift. Invited. Conf.

Procs., Photonics in Switching, Pisa, Italy, Sept 15-19.

Øderud, T., Grut L, & Aketun S. 2015. Samspill - GPS i

Oslo. Pilotering av Trygghetspakke 3. Bruk av GPS for

lokalisering av personer med demens. SINTEF-rapport

A27121.

Responsible Development of Self-learning Assisted Living Technology for Older Adults with Mild Cognitive Impairment or Dementia

209