Yes yes you want to tell people that

you're welcome to Larry we're missing

or it's OK if you're her girl.

Can you tell us.

What you saw or over here.

Realize that that I know for me.

Wait no I just thought it was or

one would just analyze You know what if it

was all very well.

Because they always

were.

I would call me perhaps two or three.

Just because of

some of them to actually

hear you more resources

than any I mean this is

what is written on the number that

I call my now friend John John found or

today in Brian Jones who's director
of the where research meeting.

He's going to present what our work.

I mean I think where home.

What are the areas where they are they
seem you know we might tell me so

I'm right.

Yes Yes OK I was saying you know things
where everyone is waiting on me.

I know you're not out right.

But I do know that I never
thought about it that way.

Yes Or before we met but Brian.

Yes

your life

you were with your.

What were your you know any you know

well.

Right.

Or.

OK All right.

Thank you.

So I'm Brian Jones On been instrumental
in the organization of this group along

with Claudia and John Sandford and Rick
Dooku is actually left Georgia Tech sense

and have been director of the Aware Home
Research Institute for the last well since

two thousand and eight but involves
since about two thousand and five.

My main area of interest is technology's

keeping track of technologies
out on the market bailable for

monitoring of older adults but
home technologies in general.

So how can we actually this
is probably my next slide.

So I'll just bring that up.

Worked over there and mouse is the last.

Area.

Of the right.

So just I'm going to go through these
slides fairly quickly because I want to

get to the meat of my questions which
is about test beds that were starting

to form so and these test beds
are from the idea of how can we

use a home that we have on
campus called the Aware Home.

Beta sites that we're
starting to develop it.

Other locations like was Leewood center
and then homes out in the community.

So I'm going to stage
the presentation in that way and

then describe some of the places.

We're coming from in our research for
the Aware Home research initiative.

So the initiative is an interdisciplinary
effort we have a number of faculty

from around campus some who
are very well involved and

have been since one thousand nine
hundred eight when we got started.

Others who come and go based on student
interest or research interests.

The list up here is which
is right into the light and

probably really hard to see a list of all
the groups here on campus in the the blue.

Text and then some of the groups we've
been working with on the right there for

externally so
people like Wesley Woods center

Emory University the V.A. Medical
Center Children's Healthcare of Atlanta

national current return communities
a lot of the means around the area.

We've tried partnering on some grant
proposals of some of you in the audience.

From the service provider side of things
so pretty broad range of interests

from around campus and
around the community and I'm just

representing a little bit of my research
and a lot of everyone else's research.

So we had a lot of focus on aging
in place from the beginning.

That's really been our
are writing about or primarily

Wendy Rogers research has has driven a lot
of that was one of the main reasons for

establishment of the Aware Home
research initiative so researcher is as

an overview we have health and
wellness at home and one of the primary

focus is there has been older adults or
chronic care of some sort and

that can be span ages could be
autism it could be diabetes.

Asthma and of course in any
kind of aging related decline.

We're looking at solutions for
self management so

I'll give you a couple of examples of that
of health and that's again spans the ages

social communication which has been
primarily geared toward older adults but

it is potentially a solution that
anybody could use and then health I.T.

or health information technology
which is our connection technologies

not technologies but connection
of individuals to their personal

health records or which can connect
then to electronic medical record.

Being used in other different practices

other areas that I'll just mention
briefly sustainability energy

resource management in the home water
etc helping people understand and

hopefully change behaviors around
that digital media and entertainment.

So really looking at what
the future of communication or

pipes to the home will be and what
the applications will be that run on that.

And in the future tools for

the home which really enable of these
three areas at the top of the list.

So if we can.

I'll give an example in the in the slides
in more detail but if we can leverage.

Inexpensive sensing that cover the entire
home with one module one plug in module.

What can we learn from that.

So that's going to be one of the concepts
I want to think about as we get a little

deeper into this the question a test and
tracking of health information.

So these are going to get more
into the well they're my.

My When P.
outline I forgot to finish up but.

So what I hope to do is give you a quick
overview of some of our projects just

generally speaking that are aging related
and then get some of the ones that

are more health related but certainly
have up in the test bed space and

I'm not going to say that in that
the projects that I show in the beginning

are not relevant because
they certainly are.

And it's a research that we are learning
from and will apply toward the concept

that the secondary I'll
focus on is what I mean

by going in a little more detail about
the Aware Home about the beta sites and

about the larger population of homes who
are actually sensing with residents and

then I have questions that I'll take
from you along the way and also.

Questions that I have for

you at the end that I'm hoping will
get some good discussion around.

I don't know exactly where
this is going to end up but

I have some some questions that
I want to get answered but

I also want to get just
general observations concerns

perspectives how can you get involved
that sort of from from the audience.

OK so before I get into this.

I want to understand the groups that
are in the room here and I have met most

of you and I try to capture some
of these up here on the board.

I know that we have folks
from industry in the room.

Some are working with devices
some are working with software.

How many of you would
fall into this category.

So about five or six.

And you may fall in sort of my horrible.

You may fall into multiple categories so
feel free to kind of self organize.

I know we have research is in the room
including myself I deal with technology.

Claudia who has left deals with design
anybody else fall into those areas

construction and.

I'm open to any suggestions.

So I just want to get a sense of what kind
of groupings we have in the room here that

I might be able to think about or how we
would split people up if we tackle some of

these questions government organizations
I know we had at least one two three.

I missed was that for here.

I think we had one research and
yes sorry didn't count what

here the research is again researchers
thought yeah that's clinical swell.

That's a good point.

I don't have that up here yet.

So I was about I think I
was five including myself.

And government was about three I'm just
getting approximate I just want to

know that we have some sampling of these
different groups service providers.

OK And so
when you when we say service providers

that could be broader obviously there's
service providers who are offering

aging services service
providers who are offering.

Like record you consider
yourself an integrator or

a service providers like OK.

And then we had some
representatives from communities.

This was raise your hands again for
the service for orders for three four.

Right.

And then communities for
the last one natural occurring.

Etc.

OK All right.

So that gives me at least I know I've
covered most calories anything else that

I missed.

Anybody want to throw up and
a grouping that might make sense here and

I'm not saying that these make sense.

I'm just.

These are areas that I that I heard as
people are listening off their their

names and.

OK All right.

So that's that's great that will
stay up there from it and so

we get to the questions at the end.

OK So when

the Rogers Center for Research and
research and education and engineering.

On Aging and technology and Hansen.

Have done a number of projects that one
slide does not do them justice but I'm

trying to summarize very briefly what we
do with where home research initiative so

they have been.

The core of a lot of our research they've
they're looking at what does it mean for

older adults to interact with technology.

What house that technology be
designed will they accept it.

What kind of considerations during
take into account like privacy

trust numeracy understanding of graphs and

such when considering health
problems robotic assistance and

I know I'm not going down the list
in order but you get the idea.

Robotic assistance when you consider
the interaction with robots human

computer interaction is
human robot interaction and

one that they had which was
a human automation interaction.

If you are provided with
automated things like G.P.S.

or similar something that the coaches
you along the way how do they.

What are their interaction with.

And in general aging in place and
the human factors.

Related to all of this
when anything then yes

you know it's

a pretty broad It really depends on.

I mean most of my focus has been on
the independent living so getting

at sixty five seventy and older or some of
the communities that we've been working

with I know that Wendy has students who
are looking at even fifty and older.

It just depends on the studies.

Would you agree with that over sixty five

but you do some comparisons of younger and

just sixty five.

I've seen that graph and
it's so depressing.

You know that's a good point.

And that's actually a very good point
because some of the technologies that

we're working on are would only be
applicable to those who haven't really

adopted cell phones or similar types
of communication technologies and

such and others.

It would because it's healthy.

It could start even
younger than sixty five.

So it just depends.

And that's why I know that Beth Minaya
who some of her research is in

here geared toward self management
of health conditions she's

looking at like the fifty to sixty age
group when it comes to diabetes but

it's still relevant to what
I'm talking about today.

Create a center first this

technology in environmental access many of
you know John Sanford there focus has been

on how can we make communities safer and
more livable from a design perspective.

Specially the home modifications devices
that might go into the home like

the the tub transfer bench or grab bar for

transferring from toilet
to some other mode

of mobility or assistance device and
then medication containers.

These are just some of the examples
of current projects that he'd give me

medicine medication containers that may
be easier for older adults to open or

to understand when they need to take
certain medication etc which is

this is an ongoing research problem that
I haven't seen a perfect solution to but

I don't know that there
is a perfect solution.

What else.

That's pretty much covers is there I know
he's working as a Regional Commission and

such and doing some of the studies of.

How people are performing in their homes.

How do you modify the homes to make
them safer for the individuals and

such cognitive gaming is one
of the projects that one of

Wendy's former students that now at N.C.
State and my colleague Mary Beth Gandy in

the Interactive Media Technology Center
are working on to understand how games

that claim they maintain cognition
see things like Brain Age are similar

really perform with older adults
in terms of maintaining it.

So and then coming up with guidelines for

how what what is important in
these games and what isn't so

that they can then build their own
game and test out how well it.

It appears to these guidelines and keeping
people in truck with or from their.

Maintaining that cognitive state.

So trying to reduce the potential for
decline there.

One of the things that they're trying
to answer is are the claims valid that

games can provide that.

So this is certainly a research
area where the big question mark

clock reader project who actually works
here in the health system Institute

as well as codes of architecture and
college of computing has a number of

class projects that start they
start out as class projects and

if they have a lot of potential

may progress to something
like a sponsored project.

Usually there are two or
three that each year coming

out of her happy healthy home course
which is offered this semester.

She'll have two or

three to really look successful maybe
some of the industry folks who come

there open house in April or others may
find interest in them and help them.

Planned out in some cases
it's been totally changing

the way that the technology is
where the technology is focused.

For example changing it from language
to understanding to something that

helps a person with visual impairment
to identify an object in the home.

I didn't have that slide up here but

that's one example that I know came
from her happy healthy home class and

is now a sponsored research project and
helping with the visually impaired

identifying objects at home or
in the in the grocery store.

The clock reader project is
one of those projects as well.

And this is getting at the I think there's
a mini cogging Samar's somewhere that

uses a clock drawing to
determine on paper to determine

a person's level of cognition or

I guess where that where they stand and

I'm sorry I don't explain this really well
it's not my my area but what they're doing

is trying to automate that where they
can track the drawing of each stroke and

understand better and possibly
automatically what they're how they

drew that clock instead of just looking
at a piece of paper and seeing that.

OK that they've drawn
it this is the result.

So it provides a little
more understanding and

then the analysis of how they drew it and
tensely that automated

detection of early detection of of
Alzheimers dimension of some sort.

Another area that we've been focusing
on as more of the communication

one that we started on was actually
present religious back in two thousand and

seven we came up with
a concept Claudia role and

I came up with a concept of a device that
would connect people within the community.

We had gone out there to present
a village and saw a gap people weren't

interacting with each other very well and
they had some of the people we interviewed

said well I just don't know who's really
here and what their interests are.

How do I get that information.

So we came up with this concept of a
device that would be in each person's home

and a default state it would be showing
pictures from the left side it would be.

Actually the left side it would be your
personal pictures on the right side it

would be the community pictures and in the
idle state it's just acting like a digital

picture frame showing those pictures but
you also have the option

of taking your personal pictures and
sharing them with the community or

grabbing community pictures and
putting them on your personal side.

Seeing who shared them and

linking up with the individuals were
never built out as a prototype.

But it's something that we certainly
have interest in revisiting calendar was

another area we looked at if you
understand the events or can change out

events instead of planning on a month by
month basis if you can change out events

more frequently than it gives the up
change to for people to engage in

kind of off the cuff
spontaneous not spontaneous so

much as you know it's more not I'd
like to present my travels to Italy or

somebody asked me about talking
about my quilting experience for

some point that gives them a chance to
kind of plan out within the community

structure beyond what the community
planners are able to do and

then messaging it within the community
which would allow them to quickly say

the bus that takes you have to
schedule two days in advance

of schedule two o'clock today to go to
Walmart if anybody wants to join us or

we're missing a fourth player and
whatever.

If anybody wants to join
us will be in the hall so

it's close to getting people more
engaged within the community.

Yes.

Funding it was it came out
of a class project we had.

We actually contracted to not contact
we hired a student to work on it and

he went on his own.

Out.

Despite my point him back and

we ended up with something that
just wasn't worth sharing.

So it really needs some some good funding
to take it to the next the next level and

start testing it out within the community
doing the proper evaluation along the way.

So it just means going after
a research project of some sort or

industry sponsorship or similar.

And then taking that to the more personal
level we had a class in two thousand and

eight that looked at a lot of
different types of devices for

that would go across housing options so
not just a retirement community but

also an individual and
living out in general neighborhoods and

what we've called that aging in place
although I don't think that's the proper

term because they see in places much more
inclusive of any place where you live so.

So the device in this case was
intended it was designed with

the idea of providing multiple
modes of interactions so

tangible objects that you could feel the
texture if you have a visual impairment.

You could have some sense of
where the person is that you want

to communicate with might be
your granddaughter or your son.

And then you have physical
buttons on the screen as well.

So the ones you want to communicate
with you pull out in the current

design you place it over in our Friday
reader it's behind this dialog box where

we think we're going to go
as you pull up the tab so

the people you want to communicate
with and then you can select right.

Write a message or
talk communicate through talk but

trying to get at the other
communication modes that children

adult children or or
grandchildren are using today's.

Your text messaging at the instant
messaging email communication a lot

of our older adults just don't know how
to use and it would really open up their.

Their options as far as communicating
with someone instead of calling and

feeling like they're disturbing them
at work etc So and then the last button

there is called share which was to quickly
send off a note like I'm leaving home or

I'm back home something that they can and
all this connected through a web portal.

But this is an example of
how technology and design.

Can really come together to to effect
change in the way that people do things

and then one of the most exciting research
areas currently is working with robots and

as as assistance in the home for older
adults and this involves the human robot

interaction that I mentioned with Wendy
Rogers Group as well as Charlie Kent's

group that is right behind us here in
the health care robotics lab the P.R.

two robot from Willow Garage
out in California and

the intent of this beta program was
to get in industry and universities

considering how this robot could be used
in different settings in ours is the only

project that actually has it in two
settings one here in the lab behind me and

it will move in April over to the Aware
Home where we will test it out for

three months in an authentic home
setting with older adults and

of course the programming
the coding side of it continuing to

do things such as pick up objects off the
floor recognize what those objects are and

determine how best to get to that object
how to retrieve it with the different.

And if and
effectors the reporters are similar

what the from the human can be
a human robot interaction standpoint

how should the robot in our
to the individual should it.

You know is it going to be intimidating
if it's always following you around or

what kind of the other questions are asked
her what kind of tasks should it perform

for you and certainly safety is a very
big concern when working with robots.

Even in the lab they have very
strict safety procedures so

that's one of the things that
we're we're dealing with as well.

So I think that's OK So

questions so far I'm still going
to continue with other areas.

Yes.

Yes.

So yes we have considered the i Pad And
actually that's our next

round of design although we're looking
at other options that are similar but

have features like cameras and
the things that the i Pad seem to

fall short on i Pads as a device for
older adults is phenomenal.

And as to how easy it is for
them to work with it.

I haven't done the studies myself but I
have looked at other people's studies and

I know that there are there is one other.

It's called the mode.

It's so it's a company that we saw
just training last in December.

They have a social communication
arc trying to connect

individuals in family
relationships as well as.

Clinicians and others in that it is their
plan for the future but through the social

communication interface on the i Pad and
I think that is a valid point for

certainly for people who have the visual
the ability to identify what they want.

Select visually on the screen that's
certainly valid what we're trying to do

with this other design is to provide
it a tangible side of it as well.

So getting at another another sense.

Where if you don't have that visual
ability you can maybe control it

through voice but
there's also that tangible side of

it which we've actually
found with with children and

older adults that tangible thing is
is pretty important when we did and

Mitchell evaluation of this in an earlier
stage which we demonstrate I think.

Maybe a year ago two I can't remember
at this death meeting we had

blocks that you could pick up in the some
of the older adults we evaluate it with

we're picking it up and and looking at
their family member and kind of holding it

in their hand and while we're asking the
questions about it so it and one of them

still golly I think I could talk to that
when you're asking what's the function.

Secondly I think we could talk to that
that device and pick it up and talks and

feel like I'm really communicating
with that individual.

So it's interesting what as you start
considering the different designs and

adopting technologies that
are already out there.

What can we do to improve upon that
yes the i Pad is our next approach but

there are some problems in extending
the i Pad to the more tangible so

to provide a more tangible interface so
we're thinking something Bluetooth that it

could dock into in one state or you can
pull it out and just use it as an app.

Yes to.

Yes So

it was

also

very.

Yes I

did my

selections.

To make.

Yeah I just write

more about.

I think just say

you know I

just I think they are marketing.

Phenomenon.

I was putting it.

Edge Ariel he has a tab a Samsung
tab He's a visiting researcher from

Samsung South Korea.

Yeah.

OK so

I'm going to keep going because I
see that we're getting shorter on

time than I had expected.

One of the other areas
that is very important

to us is the understanding of how
certainly where how these technologies

that I described would fit in
to older adults homes or lives.

But also from the understanding of health.

We feel that as we as educators we need or

academic researchers we need to stay
out ahead of the curve to be able

to tackle projects that industry and
others wouldn't be able to do so.

One of the areas that we're focusing on
along with some of the universities around

the country and around the world actually
is on the home health monitoring and

particularly of chronic conditions now
what's on the market today are usually

addressing some of the immediate
needs of the individuals so

we need to know that there
are active are up and about today.

So we have activity monitoring using any
number of means of motion sensors is

the most common medication monitoring
there are a lot of different solutions

depends on the individual and

usually medication monitoring has
some kind of reminder to take the.

Take the medicine and then some
connection to and this is from my.

Perspective so I know that there are a lot
of solutions that don't have that

connection but
some kind of connection to a caregiver or.

A web portal where family
members can keep tabs

medication is probably one of
the most important areas that we.

I think we need to find some
better solutions for but for

the moment there are some products
out there social contacts to Vittie

we've already talked about some of
the social connection aspects but

there are products out there are such
as Grand care big screen live.

Paul mail and others that are providing
additional interfaces for

I mean easier interfaces for older
adults to communicate through e-mail and

these other communication
was that we use and

it's very important to keep them and
gauged feel like they have a purpose but

engaging in communication with family
friends and others vital signs

number devices out there on the market
that are telemedicine type devices they

monitor blood pressure.

Maybe temperatures certainly weight.

Pulse oximetry even the E.K.G.

in some cases and
it very important in all of these pieces

put together can bring a lot of
meaning to an individual's health but

traditionally with the exception
of a company like grand here and

similar there hasn't been this nice
connection of all these different pieces

of glass and I have on there is
exercise and sleep patterns.

I actually am and wearing a Fit Bit
that I've been wearing that connects to

a website gives me activity
information throughout my day.

I know when I took what how many steps and

I can look back to my dango golly
I was at work that day wasn't I.

And then there are other days
where I play with my kids and

I don't realize how many
steps I've taken a look.

And so well over ten thousand.

Kids are great.

But it's Anyway having that kind of
information understanding when somebody

exercise when they took their medication
when they were communicating with others.

How long they slept that that's all
important from a health perspective and

what we're trying to do is pull all that
together and get a look down the road and

say OK if we have this long term
understanding of these different health

considerations and I'm not saying
a just this list up here is is.

All we should be considering.

But if we look at health in general in all
these different areas that we could be

monitoring automatically.

What can we learn what can we.

What kind of feedback can we give to the
individuals or to their clinicians and.

The like.

Or.

Well more or less

right right right right.

So.

Right so you're getting exactly
at the point that I want to

the test bed idea that we're
coming up with this and

you guys are already out there doing
your testing in the communities and

certainly if those kind of
partnerships are exist.

We'll be you know we're open to and that's
one of the things I want to hear from

from the audience today as we get to the
questions at the end is how could you see

yourselves being involved and
it may end up being a homework assignment.

But let me move on and describe some of
the things that we're considering that

from our research that
one of the question.

That last list.

Hopes.

That's not it either.

They're.

Right.
So so

the idea is yes there will be some kind
of platform and I know grand care already

offers something like that that third
party vendors can plug into one of

our concerns as we go to larger numbers
we need to look at open source.

So we were looking at some of the options
that are out there on the market that

would allow for pulling together all these
different sensing technologies plugging

them shipping them up to the cloud and
then in the cloud analyzing them for

health trends like the example
that I've been working with most

I'm most familiar with Ted Johnson Emory
University helped with which is nocturia.

How many times is somebody getting up out
of bed at night to go to the bathroom

can be very important for them in picking
the right medication for them and

understanding that when did they take
a drink of water which is very hard for

us to figure out when do they go to bed.

How many times when did they kind
of wake up one of the stories.

That's OK It's not that important but
one of they get up out of bed.

What's the path that they took so

from considering the safety of
that pathway when they got to

the bathroom did they turn on the light
which can affect their weight state.

How much should they urinate he's looking
at the stream the strength of the stream

things like that which again are hard for
us to capture at this point.

How long in the bathroom how long
did that it take on the go back and

get back to sleep and so
we had we're looking at pieces of this and

some of our prior research but
I'll get to in just a second but

trying to get something that we could
provide to clinicians or as feedback to

the individual that can change behavior or
change the personalize health more so

if you know when they took the medication
in addition to looking at a month

worth of a data of how many bed exits
they had to go to the bathroom then

the clinicians say OK I want to try
changing a medication changing a dose and

changing this set of the other you know
make sure you don't drink water after

six o'clock I don't know different things
that could affect their their health.

From the technology

side we've worked on the initial
steps of this but it's just to prove

a I'll get to this in just a second
I can show you kind of the framework

we've come up with so let me step
through some of these other slides from

a perspective of the low cost and
and grand scale.

Some of the policies that we've
developed over the last three years or

so and that have continued to be
developed at University of Washington or

one of our students is now professor
under this was under Gregory about

who some of you may know was director
of the health system Institute and

till this month they focused on if we
have a single point of sensing if we.

Can plug in a module into the wall

that can monitor the noise of
electrical signals on the line.

What can we learn from that that will
basically every signal is the is unique to

the every appliance has a unique
signal on the power line.

So we can once we know what each appliance
looks like we can start tracking.

Trends of activity in the home just so

that plug in model instead of multiple
motion sensors around the home.

Now I'm not saying we get rid of the
motion sensors I think we have an initial

research phase where we have to test out
how does this correlate with what we're

getting from motion sensors water
sensor can you screwed on to the end

of a spicket on the outside of the house
and open up the spigot and then any time

a water device or faucet or washing
machine dishwasher etc turned on it.

We can tell which one it was turning on
and potentially how much water they're

using because of the pressure
change in in the line.

Gas Works the same way.

And so the idea was to get
some kind of understanding

of whole house activity that can
provide as an environmental cues.

So one we can track back to the any of
the individuals over time and look at.

Well OK this is change what does
that mean could be as simple was OK.

They are they are preparing
coffee in the morning or

they haven't been running the microwave or
the.

The than with the stove top and
they didn't open the refrigerator door

all these things have to have
electric signatures that could

help in identifying some kind of
behavior that would be important to us.

So this is kind of whole
house sensing it certainly

in conjunction with vital signs and
things like that will be important.

What we're looking at is if we
look at the long term things like

facial details might change so

we might be able to pick up on hotspots
under the skin in this case we have.

It's called the intelligent
bathroom project this

Georgia Tech Research Institute is working
on this three components the smart

floor smart toilet and it's and a smart
mirror with feedback so on the mirror they

have cameras mounted that can look at the
infrared complection of the face pick up

on on heat hotspot areas as well
as a visual picture of the face.

So maybe you can't see it visually but
if we capture.

Pictures over time.

Look at my saying I want to do this.

Look at my picture six months ago or
a year ago compared to now.

And notice some of the hot spots in
the infrared that have maybe shown up or

gone away it just depends.

And maybe that that is a trigger for me
to go and get talked to dermatologist or

some other kind of health concern that
that I should talk to a physician about.

So the mirror display may it blends
into the mirror when it's turned off but

there it would cue you to
do certain activities so

one thing it might do is ask you to
track a moving object on the screen.

We can potentially and
this is Ted Johnson.

Potentially track or determine someone's
I can't remember what it was but

based on how well they check that object
there's some health condition you can pick

up on and so and then the floor
is looking at balance looking

at weight changes potentially of course
if we have a scale that would be nice but

having a built in the floor is even better
the intelligent toilet was intended

to use a sensor forty texting certain
chemicals and apply it to the toilet.

One of the most valuable.

Of course Ted is really
excited about this right.

Being a incontinence and
everything you're unary and everything.

He had all kinds of ideas about how these
this combination of sensors could be used.

One of the things that we could possibly
sense is that they actually have taken

their medication because traces of it
would show up in that in the output and

I think there's there's
a lot of potential there.

But a lot of research to go on
before we get to the point of

the right

place here.

There is a lot of your

life.

There was

no I don't know if you will burst

your letters

were

always.

You

know you're right.

So.

Mine's more silly.

Yeah I think that that really gets said I.

Obviously partnering with other people
is going to be very important to

absorb some of the costs
of of such an endeavor but.

Right now or

have twenty different products that
don't work together and right.

And that's exactly where we're trying
to go with this is trying to get some

commonality between the sensors and
I mean the same sensors can be used for

sustainability monitoring of energy
management they can be used for

digital entertainment media and

that might be actually the way that
some of these get into the home but.

Right.

Yes.

The from a from a health system
from a health I.T. standpoint yes.

So from what I've been talking with
some of the other researchers here on

campus who are trying to put together
health information technology test bed.

There's there are formats for
devices to share information with

the health information infrastructure and
that's something I'm not an expert in this

area but that's something that we will be
considering once we get to the point that

we can we can look at all this and
go OK when we start tackling this problem.

These are the these are our partners
this is what they have to have this

is what we're going to do on our own.

And it is I think what we're
mostly focusing on is can we get

off the shelf products that work together.

And where we can look at more
of the analytics side of it.

To take those products what
the information from those products and

look at the holistic picture.

So what are just bits and pieces like
I talked about on a couple slides ago.

Can we pull those things together and

can we fill any gaps sensing
that's necessary to which.

Where I was getting at with is this
some of my questions are going to be

what are the conditions were looking for
what.

How do we identify some
of these conditions.

You know one of the some of
the pieces that we talked to with

with Ted Johnson of course I knew this
was going to be longer than an hour and

a half I knew it was probably
could be a full day workshop and

so trying to trying to get some kind of
feedback from the group here this is been

great talking through it because
it changes my my thinking on it so

I'm going to put press on until about
ten thirty and then we'll dismiss but

I'd like for everybody to provide feedback
to me if you have any comments or

questions or we can arrange
some meetings individually or

possibly follow this up
in a different session.

If folks are interested.

So you talked about being able to
build systems that have an A.P.I.

and can support other
products the salute is

the software that was worked on here and
both my nuts.

Was a slab.

Eugene Minsky sim I'm saying that
right has been developing it and

came up with a an application
that was kind of a motivator for

the underlying infrastructure so

the underlying infrastructure is
the most impressive part of this.

It consists of a data store that is
easy to hook into and in fact on some of

our other projects related to it
where home we have use that A.P.I.

just to get to the data store and
not have to design our own database.

Very easy to configure and set up fields
if you know what database fields are.

The application that they are running
on the system is health very much like

with the fit it provides you can keep
track of how many steps you took but

total for the day.

Not necessarily individual steps
throughout the day you can look at

what your.

Wait trends were and what you're
seeing here in the plot at the bottom

is an example of here's my weight trend.

Actually this is the other researchers
in the demo weight trend over.

I think it was like nine months or
so is pretty long period and

exercise picking up at the bottom here and
how those correlate those are the kind of

things that we're trying to pick up on and
while this is an obvious one.

There are many others out there that
are not so obvious the other thing that's

where this goes beyond what
software on the that is

you can analyze this with this kind
of comparison of two plots on fit but

you get to look at OK
there's my thirty day trend.

Where is this when you can kind of pull
variables that may have no logical.

They may not make any sense at all.

So these are days that I played with my
kids or when my child got in trouble but

here's a sleep in their sleep pattern and
look at questions that you may

have the you want to discover on your own
and this gives you the option to do that.

Being able to set up your
own your own tracking

fields your numbers that make sense
from those tracking fields like

just marking when my child
had problems at school.

These kind of things making it

that flexible has made it
useful in many domains.

Now the now is the side of this
also allows you to stack these

things on top of each other so it just
depends on how you want to visualize it.

But there's a goal setting
which is actually what

these little the little blue line
is if you can see that there's

the ability to put the plots of the top
of each other or separate them out.

Depending on what works for you but
to see and identify trends or concerns.

For how to do that.

I had a student.

To that part I believe it's based on
what protocols but I'm not positive.

It's all this is all
application on the Web So

most likely that's what they've done.

There's all

this other things that they had
on top of this i Phone app and

Android app which makes it easier for
you to enter your data but

it is all based Currently
the system is designed for

you to enter all the numbers yourself
calories pictures of your food

weight each day etc What we
started moving toward hoping this

like it was to look at automation of data

like we've been talking about for
the second half of this presentation.

If we start automating data into this data
store how and how is that data useful and

it totally blew a hole in their
visualization they needed to go down for

they only allow you to visualize it
a month and really we need to get down to

the day with a week or
getting at different values or different.

Comparisons that make sense without
putting too much data on the screen or

in our case crashing
the the display because we have put

one point five million points data points
into the end of the data store and

couldn't visualize it but
they dress the problem.

What you.
That's what's nice about having the group

working here on campus.

So anyway what we did with this
system was we had our goal was to

design a framework of how you get data out
of the home from these types of devices.

Maybe some pre-processing where you might
consider you don't necessarily need

all the raw data.

Maybe there are certain key data points
that you want to ship to the web.

So in this case the Web is
our data store salute and

then some kind of processing or analytics
which we're using Google App Engine for

and it could suck in the data from the
data store do some kind of processing and

put new points out or alert the caregiver
clinician that there's a problem or

the individual that there's a problem.

And so it's this pull and
push between the data store and

the application being able
to make sense of data.

OK And this is where our research is for
the future.

It's that making sense of
data sensors now exist for

us to capture a lot of this information.

How do we make sense that you know
from the standpoint of health for

the individual in different conditions and
then we want to show some way for

the clinician to have greater control and
I'm not saying the clinicians cells but

possibly their office to have greater
control over what they want to learn

in the home so in our case study run
nocturia that I described a minute ago

giving them the option to turn on and
off different sensors in the home.

They don't have to know what the sensors
are they just have to say I want

that exits.

So now we turn on the different
sensors and collect that data and

run the processing the say OK.

There were three bed exits
during the night on average over

this month that way the clinician
doesn't have to take the time to figure

out can he do include the bed panel that
monitors heart rate and respiration and

needs to work in this way and.

I need this sensor in the back.

They don't need to understand that they
just need to know that this option is

available and so that's one of
the one of the areas in which

this is kind of moving into the test bed
is being able to consider what what kind

of operations our clinicians are or
the individuals or the caregivers or

whoever going to need when we
start building up such systems.

OK And
now I'm finally at the meat of it right

I definitely wanted to frame it before I
got to this point because I think that it

helps in your understanding of what we're
talking about when we get to the test but

so the where home on
campus already exists.

We already have some technologies
installed in the home.

We view the Aware Home as being the
starting point the alpha site for any kind

of experimentation with health data we can
have people live on one of the floors.

We actually had a student living there for
a month to tell me what was missing.

So that we could support people
living in the home long term.

That was actually one of the plans from
the beginning was to have people living in

the home but
it never worked out because of the.

Well just six of human
subjects seeing employment and

charging rent for in our case we're
trying to get students to live there.

So once we figure out things in
the home we polished it up and

we think we have something that's
worth testing out in the communities.

It's much easier to get a population and
community if you are have something

central to that community so we have been
working with was they would center and

the senior living towers on the campus
which are independent living.

There's one floor of assisted living but
so

it gives some some mix but they gave us
a unit in the facility that allows us

to have a location where
we can set up the data.

I mean set up the sensors
walk it up at night and

it doesn't change so
we could have somebody come in for.

Or five days at a time and
just test in that beta site.

But what we don't get is
the realistic data right because it's

not their real home.

So it's more like testing in a hotel room.

But it gives us some other answers
that we may be looking for

Beyond that the goal is to get into
real residences and this is not new.

We've been doing this with a lot of
our research projects it's just been

we've been identifying

individuals that could test out our
social communication device or that

we could ask questions of around something
design guidelines or something like that.

Now we'd like to get data.

I mean if you think typically
our projects go about me.

Maybe three months I've seen.

But beyond that usually
the research is done and

they've reported everything and
everybody goes their happy way.

Now we want to be able to look at that
longitudinal data over long term and

really look at those trends of health data
which are long term trends and the only

way we can do that is by having this type
of sensing in people's homes for years.

So what we are looking at is one

mechanism was to offer an industry a test
bed where we would go out and recruit

five hundred individuals out in the
communities that we have around Atlanta.

Not necessarily based on condition just
having a set of individuals we could call

on and monitor when I say monitor I
mean keep track of their changes in

their health just from their self report
we interview them once every two months

we know what kind of networking challenges
will have we installed some kind of

technology there and maybe we establish
the network there from the get go.

This is something I've been researching
on about to deliver the information to

Georgia Tech Research Institute
which may actually fund it.

And then industry we appeal to industry
and we start small fifty twenty twenty and

fifty homes we pill the industry and
run a quick test if it

fails great we drop the whole idea and
and we move on if it works.

We continue to build up
to five hundred homes and

then we try to keep those homes engaged
in some kind of of research with industry

industry products at least calling up
the individuals every two months and

reassessing Has anything
changed in your home.

Like networking has your provider
change has your health change.

You know what can we get from
the individuals that would help us keep.

Up to date database of those individuals.

And so that it ends up giving
us a revenue stream back into

the initial investment that we made and
the helping eventually to fund some of

the research around this based
on my feasibility study.

I think we can sustain that
after about four years and

then that gives us the opportunity to get
into these five hundred homes with some of

our research projects that are trying
to answer these more long term goals so

since it trying to buy sensing for five
hundred homes out the door is just crazy.

We know we can't do it.

We know we have to have some way of of
getting it to work and self-sustaining.

And that was one of the solutions
we came up with Ideally

and this is these are my questions to

you in the next ten minutes I'm going
to throw this question out there.

The one one question I can get out
of you is if we have this in place

this five hundred homes and
let's ignore the industry said

let's just think about the health sensing
side that we've been talking about here.

Number five.

So monitoring long term trends in health.

If we do this.

What kind of challenges or

what considerations should we be
taking account of just to is very.

Broad question.

I already know some but

I want to hear from you because
there are a lot of things.

I'm sure I haven't considered.

So again long term trends.

How should we be approaching this.

Should we be considering
starting with the communities and

because we know they're easy to get at
individuals or should we be focusing on.

Clinicians practices and conditions and

going from there and I know that the
question is when we're looking at health

that's probably most logical but what is
your feedback from your different domains.

Anybody want to.

Venture that yes yes.

So

that and alumni.

Is another one that we've considered but
yes you're absolutely right.

That's that's one of the populations
that would be very right.

All right.

Any other comments suggestions
concerns exciting.

Whatever.

And so you go.

With.

You know what.

They use.

This.

So.

Well you're right.

Those.

Were.

Right.

You know.

This is why.

The office is right.

There can have

an industry is already starting to
jump on a lot of these these issues.

What we're trying to do is look at them
from something that's less costly could

go to scale fairly easily.

But having this long term.

I mean if you look at what
Grant care already has in place

they have more homes than we have but
are they monitoring for

these types of things or
are they monitoring for like for example.

Organization is so wide.

Through and through those.

Forty or.

So yes.

Us OK.

So that's just.

Like.

Yes.

OK you're right.

That's not a.

Why yeah.

I just thought Yeah.

Right.

That's what some.

Absolutely.

And that really gets me where
a lot of the funding for

this kind of long term tracking is going
to come from helps you know helps and.

Just.

Right.
So that's a very good point and

I don't we have a lot of studies
that certainly Wendy's group has

has worked on that where they
actually maintain a database and

I know that Emory does too but of possible
subject people who are interested in

participating in studies what
we have not done is drafted OK.

Partly because we don't know yet
what it's going to be

this is what this particular
test bed is going to involve And

that's something as we get closer we will
have to do and we will have to go out and.

It really is easier to get at
some of these communities but

we may go through like the area agencies

in the different areas where we
really would like to get Georgia.

Well represented from
demographic from economics from

all these other standpoints.

But I know they're going to be a number
of challenges that are along the way.

OK So that's kind of.

So this is really the five hundred
homes that we're looking at

becomes a piece of a larger effort
that Intel has been pushing for

which we just had
a workshop on in December

to hit ten thousand homes managed by
different organizations different

universities around the around
the country around the world.

Mainly to get the U.S. back on track
with what Europe's doing right now.

But in that particular effort
they want to standardize

everything that goes into the home and
so it is a this.

It's these are the sensors that are going
to go into your particular home.

I'm not convinced yet that that's
the right way to do it especially when you

consider the costs if I don't need to
put something into someone's home.

Should I.
But

at the same time if it's something that
can give us to capture data that might

be important when an event occurs that
we could look back and say OK could

we determine early on that this person
was going to experience that event.

A fall a some kind of
congestive heart failure or

other stroke etc could we
pick up on those things and

have gotten them the help
that they needed.

Earlier on.

That's one I think it's a good argument
to say OK but let's look at putting

additional sensing in there that we
wouldn't have needed based on the current

needs of the individual but when we
look at it from a product standpoint.

Certainly it would only be those devices
that are important to the individual.

So from the research perspective I would
say it probably makes sense to have

some core set of sensing.

Have subsets of individuals you are.

Where you're focusing on different
conditions where they may

be have family history of
congestive heart failure.

So let's make sure that we have
this sensing in their home.

So I think it's really building up and

this was one of my questions today for
everybody.

What are the conditions we should
be targeting what are the.

What are the signs of those conditions and
what could you imagine in

the environment in the vital signs and any
other data that we could be collecting.

Might be valuable for

grander understanding of you know not just
what's my heart rate at any given time.

But what's my behavior on the home of
my sitting down more than I was six.

Months ago and
how does that figure in everything.

So I think there are a lot
of questions out there.

I don't know the answers yet but.

Right.

So the most immediate thing
would be to identify it.

We been looking at medication
monitoring solutions and

I haven't found one that's
going to work for everybody.

It just doesn't exist and
I don't know that I ever will.

Because everybody has a different way
that they deal with their medications and

different needs.

As far as being able to remove
tops carry it with them etc So

medication is one area that I
know is can be challenging but

as far as activity that's something
we can do right up right out the door

because it already exists there are bed
solutions already out there for

monitoring of sleep and
heart rate and respiration rate and

motion during the night so
we know those solutions exist.

Most of them are pretty expensive.

We have a what's called Global
sensor network is the middleware

that manages the in the home runs on some
kind of box we're using a plug computer

in the home that can take in the data
from these different types of sensors and

get it to the web so that I think we
could provide some kind of gateway to

the home something that's already exists
on the market or that we can customize.

So you know.

So I am actually looking at the Big B.

is possibly a solution but they don't
have enough products are very expensive.

They don't have enough
product out there yet.

Z. wave is potentially more available

little lower on cost X.
and obviously that's what

most people are using because it is low
cost but it has some reliability problems.

So I know that that's out of
the realm of most people.

It's not ready for tomorrow.

You know now.

We're actually in the stages of applying
for research grants for that whole

home sensing where we would have something
like the motion sensing that other

companies are using right now just to
compare the data that we're getting from

the whole home sensing from the electrical
the water etc with the data we get from

the motion sensors and seeing how do they
compare how does one augment the other.

So that's kind of where we stand
in our research right now but

what Georgia Tech is trying to figure
out is how do we put a significant

investment into our
research to get to the next

stage that other people haven't other
universities haven't even considered.

And obviously other universities
are considering this but

we need to leapfrog because
we've gotten behind so

this is my kind of initial
feeling out of what's possible.

How would we structure it and
presented in I think it's April or so to.

Georgia Tech administration and
say this is what we've come up with.

If they choose to fund it.

Great if not back to the drawing board but

it was there are a number

of products that are certainly
telemedicine products are already there.

There are some medication like Glow Caps
that we could that I've already talked and

they said we could tap into their database
of when people remove the cap but

there's some problems with the Glow Caps
themselves so their challenges

all along the way and I don't know yet
how we're going to resolve them but

I will will come up what I'd like to do is
say just like we've done with nocturia So

these are the things we want to know these
are the things that can't be solved right

now but we can move toward that.

But right now we can
detect when they're in

the bathroom at the lights
on when they're in bed.

What their hurry respiration etc is.

Maybe with an X.

box connect type sensor we can detect
the pathway that they take to the bathroom

fairly inexpensively but that's
research that has to be done still so

there is that so I think we're going to go
as we're going to identify what conditions

just of heart failure is very important
because Ted Johnson is nearby and

that's his expertise can be an indicator
an early sign of other problems

that may exist so that could be important
too and I think it's those kind of tiered.

Considerations not Tyrian in
itself could be an indication that

a maybe same strong congestive
heart failure is possible but

it certainly has a link to other
other failures or other declines.

So maybe that's where we should be
focusing right now is that those

early indicators of any
kind of conditions and

then focusing more on the conditions
once or experience and skill.

You know.

Yeah.

I.

Am.

I know that we have some hospitals in
Savannah area that we're very keen on

partnering with us and some other areas
here attack mainly health information

technologies but we view those as
being but I mean they were actually

want to partner with us on this on this
whole home sensing aspect so there are.

There and there's so many possibilities
for partners out there that it's

it's a bit overwhelming but yeah.

OK.

Yeah.

Yeah.

Yeah.

Right.

OK.

Your story.

Is.

You know it needs to be provided
on a different manner and wow.

So what we're

doing is of yours.

You've got to go and.

And everything in the home and so.

I did you know it's like well this is.

More you know our room.

Mention.

Some of those various aspects.

So I think that's a great example of where
the health information technology is going

to be extremely important from a personal
health record from what we share with

the individuals and how it shared with
on one of the other areas that we know.

We need to focus on and
other institutions are already is

the feed back to the individual how do you
put them at the center of their care and

really put them in charge.

You're right.

From everything I've heard from
complaints from the outside world.

It's.

Basically when you're in the fifty
to sixty something you're

taking charge of your own health and
at some point it just becomes so

cumbersome overwhelming or something
you just can't rely on the doctor.

And we need people to start asking
questions we need them to start

understanding better.

What it is that they have and help them
quickly get access to that information and

not additional information
that is invalid.

You

know he's.

Yeah and

that's where the foreign pharmacies are
trying to focus now but they can't do it

because different people go to different
pharmacies for different purposes.

And it's interesting.

Right.

Here.

You know.

And the other thing that that does.

Yes that's a very good point.

The other thing that does for
us is essential lies the service issues.

So when you start talking
about truck rolls.

And you know truck rolls as utility
companies have to do whenever power goes

out for somebody.

It's a huge cost and as you can
imagine with technologies like this.

They're going to fail and they're
going to be responding in some ways so

having things in a central location and
having that ability to interview multiple

people while you're out there is one of
the things I've kind of based some of

our our feasibility studies on
from the industry perspective and

that's kind of where I
thought I'd focus with this.

Grander health trend research but.

I'm glad you mentioned the nurses or

on campus because that's not
that's something I thought people

were probably going out to external
facilities for their care.

But.

And yeah I think that's
a great suggestion and

that makes it feel better because I was
looking at the challenge trying to get it.

A number of homes just
everywhere scattered around and

it's just over one hundred.

OK Now I talked to the one in
Louisville that was very keen on

this idea they have four hundred residents
in different types of settings but

it seems that the continuing carry turn
communities are the ones that offer those

kind of services and are more motivated
because of the cost of keeping them

in their homes versus in skilled
nursing or any other system of.

OK many other comments son.

I know are way over to.

But I really appreciate everyone sticking
around and and providing your feedback.

If you have any any questions any
other ideas feel free to email me I

have a stack of cards I can
hand out of people interested.

Thank you.