One thing I do have to say is you
guys will be looking at me and

the slides during the presentation what
I'm looking at right now is the food.

So if I start drooling you
are going to insulin shock

something like that and
you'll know why I'll be staring.

At the food for about the next though.

You guys are really
putting on the dog but.

But the downside of that is I
would be looking at the food

in the whole presentation.

So the title to the talk
today is imaging in

healthy aging and
dementia bigger sandbox and.

The title is inspired by recent.

Trends in the all timers
disease world and.

You know those trends essentially our talk
a little bit more about this in a minute

that all timers disease and aging are
merging as the all timers folks begin to

look for biomarkers of all timers
disease in healthy aging folks.

So we're all going to be playing
in the sandbox the same sandbox.

Already are in fact and.

That trend I think is
going to continue and

I've become convinced actually in
putting this presentation together

that it's really going to change Aging
Research in some very significant ways and

you know the ways we get funded and
so forth and

Karen I study sections and so
on will probably be changing as well.

So.

Let me first.

Get my mouse to work.

OK Well yeah obviously I've co-opted
the mouse with laser pointer but I.

Want to mention my collaborators
on the first forward to change

to ping who Jason Langley and
Lisa Krishnamurthy.

Are all in more physicists and

are largely responsible for
programming the pulse sequences and

also I had some input to
picking the pulse sequences

along with these folks that we're
using in the Healthy Aging study.

At Emory or what we call the Emery
brain imaging project but

you'll hear me refer to it as the Healthy
Aging study because it's a sub study of

the Healthy Aging study and
our post-docs an image analysis folks are.

He cheesy Anglin and

twang honed by a hand and
a stage of both Ali So

it's a fairly large group of people
that we're talking about here in and.

They're the geniuses behind
the pole sequences here.

So what's going on with all Simers disease
and I think a good example of what's

been going on one of the things that's
been going on recently it's motivated.

Some thinking in the area
of alzheimers disease is.

The failed clinical trials with
the early all Simers disease and.

What they set out to do was to change the.

Typical all timers
biomarkers of amyloid or.

Using monoclonal antibodies like
Bappa Newsom Amber Solar News a man and

as it turns out they were pretty
successful in doing that this is from.

Clinical trials for all timers disease
presentation of Fox in two thousand and

twelve and they're showing here
that with a four carriers.

The genetic variant of
the apolipoprotein in.

That bapineuzumab could
significantly reduce

fossil OTOH in the cerebral spinal fluid.

At this five milligram
per kilogram dose and in.

For non carriers they need a little bit
higher dose to achieve the same level

of reduction and
these are changes from the mean in

standard error of the of the groups and
so as you can

see the changes in terms of the Standard
years are actually fairly large changes.

But the reason why the trials failed

was that the cognitive in
points did not change.

And one of the.

Thoughts about why that happened
is because by the time you get to

early Alzheimer's disease when
your patients are symptomatic

that the horse is already out of the barn
the damage has already been done and

you can't change cognition and
this is prompted people in the all Simers.

Disease research area to think that if
we're going to intervene in all Simers

disease successfully what we need to do is
to do it before people become symptomatic.

So that means they're cognitively
normal essentially and

you are looking for biomarkers both to
tell you when you need to intervene but

also to tell you how you need to
intervene or what you need to do.

So what does that mean

we're all going to be playing
a bigger sandbox as I said earlier.

So.

One way to conceive all Simers
disease is as a cascade of events and

this is from an article by
reason Sperling from Harvard.

In two thousand and eleven and
she proposed this cascade in

the earliest of the markers
of the things to appear

is the abnormal accumulation amyloid and

you can measure this either in the
cerebral spinal fluid or by PET scanning.

And.

Amyloid you know occurs in
about thirty percent of

cognitively normal folks over
the age of sixty five and

its star it starts off around
it you know in sixty five around

fifteen to twenty percent something
like that and then becomes a.

Positively Excel orating function of age.

So it Excel or it's the older you get so

the idea is that in your preclinical
stages where the cognition is still normal

that you can get a fair amount of
amyloid communiqué mutilation and

at least according to this figure
at me even peak before you even

become symptomatic and
get mild cognitive impairment.

Of the next thing in the cascade at
least of known elements is the Sinak to

dysfunction which is basically
glucose metabolism is measured by PET

scans closely followed by abnormal
levels of Tao or FOSS the Tao again.

Until recently measured
primarily in the C.S.F. but

there are biomarkers under
development to look for.

Deposition as well and

then after that you can begin to
see changes in brain structure.

For example a good predictor
of conversion from.

M.C.I. to all Simers diseases
size of the hippocampus.

And then after that is cognition changes.

And you begin to show cognitive
changes of course that in C.I.

that get worse until you get to
the dementia phase when you begin

to see changes in actual daily clinical
function in the idea that mild cognitive

impairment is you have the cognitive
impairment but it doesn't affect your

activities of daily living or
instrumental activities of daily living.

And when it does that's when you
get to the to the old Simers stage.

And so one of the things I think
that we're doing is searching for.

Biomarkers even further back is one
of the things that we want to do so

that we potentially could even
prevent the amyloid deposition.

And what that means is
that all Simers disease

research is trending towards
looking at healthy aging.

And indeed I think you know if there's
a take home message from the talking

you'll see this in about three or four
of the studies I'm going to talk about.

It is that.

Our our world and
aging research is going to be changing.

And I think word of stage now where you

know if you want to be one of the pioneers
you can help determine some of the ways in

which it's going to be changing but
it will be changing in terms of what we're

looking at and what it's going to take to
get funding for me and I and so forth.

And indeed as you saw on the omnibus
budget bill there's more money for

all timers disease and so

forth the I guess a lot of our
Congress folks are getting older and.

Beginning to worry about their.

Preserving their cognition.

So I'm going to.

Well let me go let me just go through
a little bit more background information

first and then we'll start to talk
about some of the imaging your imaging

biomarkers that we're going to use in our
study so one thing we started to do when.

When the Healthy Aging study
at Emory came into being or

the Emory Healthy Aging study you'll
see me refer to it as both and

again really what I'm involved in
is a neuroimaging part of that but.

There's too many acronyms involved so
I'm just going to refer to it as

the Healthy Aging study we began to look
at other large studies that were going

to be done the Healthy Aging study
is a study where we're trying to

in an internet survey collect data
on one hundred thousand people.

And then to do a deeper dive
in the King at biomarkers.

Blood base biomarkers C.S.F.

biomarkers genetics to know
mix proteomics you know etc

cognitive testing and so forth in about
thirty five hundred people and now those

thirty five hundred we're going to try
to get imaging in a thousand plus folks.

So it's.

You think that's
an ambitious study right but

we began looking at some of
the other large studies that have.

Are are being done longitudinally and
mon is

the German national Cold Heart Study where
the goal is to develop new strategies for

risk assessment early detection and
prevention of common wind spread diseases.

And.

Focusing on the emergence of chronic
diseases including all its arms and so

forth they're going to have two
hundred thousand study participants.

Aging looking at aging between
the ages of twenty to sixty nine

years old in eighteen different
centers where thirty thousand

having him are I skin hands and the M.R.I.
images are going to be the typical T.

one weighted structural imaging flare and
resting state.

Bold F.

M.R.I..

So that's a lot of subjects.

There.
Another study in mostly in Germany called

the Rhineland study on the goals
are identifying modifiable and

non modifiable causes of neurodegenerative
or neuro psychiatric diseases and

looking at biomarkers for
individuals at risk they're going to have

thirty thousand people thirty
years of age and older.

And seven centers and
their core imaging protocol is

a little bit larger than the German
national study they'll get T.

one in T two multi eco T two star rated

images which will you know you can
do quantitative susceptibility

mapping from functional M.R.I.
and spin eco E.P.I..

Diffusion waited him or I and some of the.

Each participant undergo at least one of
the other imaging modalities including.

Body fat evaluation metabolic
imaging quantitative.

Imaging magnetization
transfer quantification and

high resolution images to look
at the hippocampus in locus or

earliest and so forth so
again another thirty German seem

to like the round figure of thirty
thousand for their studies for imaging.

And that's kind of dwarfed by the U.K.
by a bank study which.

Is going to be looking at one
hundred thousand participants

launch a tune we between the ages
of forty to sixty nine years old.

And within the image cohort They're
expecting about one thousand five hundred

participants to develop Alzheimer's
disease by two thousand and twenty two so

that leaves them a fairly large sample to
begin to look for predictive biomarkers.

And M.R.I.
images are listed as T one T two flare

susceptibility weighted M.R.I. wresting if
I'm right to ask if a Marine diffusion F.

M.R.I..

OK So looking at the world of large aging.

Old studies.

We came to the conclusion that we're not
going to be unique by collecting images

in a thousand subjects in a world
where thirty thousand and

one hundred thousand are the number of
subjects you're going to collect right so

one of the conclusions we came to
was well I think what we can do

we rethink what we can do is to do
more candidate imaging sequences for

biomarkers and
then narrow down our imaging sequences by

pilot study and that's where we
can be a little bit more unique.

And so what I'll be doing today is talking
about the thirteen poll sequences and

some data supporting from the selection
of the poll sequences that reusing.

The studies chosen I mean this could be

thirteen different lectures you know one
on each poll sequence and what it is and

and the data supporting
our selection of it.

But I've got an hour to talk so
I chose some studies and.

They're good examples is the reason
that I've chosen them but you might ask

good examples of what and that's where
there's a little bit of diversity

some of the studies are a good
example of why we chose the biomarker

some of them are a good example of
how not to analyze your data I think.

And others of them are probably
the ones that look like they are going

to predict our future and
studying cognition and aging.

OK so.

Yes you're.

Well yeah so

you know how to answer your question if
they were to go out another five years and

I don't seem they're going
to stop in two thousand and

twenty two they just picked that date and
told you how many they expect.

The you know the biggest risk for

all timers disease is age
yeah that's right age.

So the longer out they
go down the curve should

be exhilarating in terms of the number of
persons out of their hundred thousand.

That have all timers disease so
they go from.

You know that figure well I think
was given around twenty seventeen so

if they go another five years
out you could expect to have

more than another eight hundred
participants with Alzheimer's

disease because it's a positive
you know the people who get all

Simers disease eventually is a positively
Excel orating function of aging.

OK.

And I don't assume they're going to stop I
mean why would you give up a database like

that I mean they're still studying
the Framingham Heart patients that started

like quite a couple decades ago or
more actually and

they're still studying that cohort
is becoming particularly popular

cohort with the also members community
because they have such long longitudinal

data why would you give a you know why
would you go to the trouble to do this if

you weren't going to fall in over a long
period of time which the implication of

that I think is that you know it's going
to involve a generate a generation or

two of researchers right were some
of the older guys are going to be

handing off the baton and
probably some of the more image

interesting data may come out twenty or
thirty years from now maybe.

Looking at the age of the subjects so

we have think this is
necessarily unique but

we have a conceptualization of aging and
aging related diseases in our lab

that I wanted to mention to you
very briefly because it'll.

It's my bias and.

So it probably will color some of
the things that I talk about today.

And that is that there's
an interaction between aging and

aging related diseases so
this is just made up

curves they're not representative
of anything in particular

other than you know you look at cognitive
measures probably a better word for.

These values over here and we know if you
as you get older your cognition declines.

But the idea Sheer is that
as your aging inner X.

with a.

Some disease process that it's going to
do it in a way where there's an aging by

disease interaction and
the older you get probably the more.

The disease is going to take
its toll because just simply

of the fact that you know you're
losing your cognitive function.

As you get older.

So you know how does that.

How does that translate
into something clinically.

Well one of the things that
happened in your phase your world

in the early two thousand as Marnie Naser
out of Boston published a study

using R T A mass in the right
parts triangulator us and

the idea was that if they inhibit
right par strain heiress and

they knew this from some
pilot data that you could

actually improve word finding or other
language functions and folks and they were

successful in doing that it's been
replicated at this point and so forth.

So there conceptualization
was that in a phase except

Jex there is an over activation of parts
of the right hemisphere and if you

treat that over activation basically you
can get improvement in language functions

well as you may know in
a lot of cognitive functions

that are lateralized you
get more activity in older

folks in the non-dominant him a sphere for
that activity and so we've shown that.

When you.

Well let me get into that in
a little bit later because that's

part of our research I think I've
got several slides on that so.

To talk about the Healthy Aging study and
the neuroimaging

part of that in terms of neuroimaging the
types of things that we're looking at are.

First.

Endothelial dysfunction or

other words that's the wall of
the blood vessels right to.

Get more torturous as you get
older cerebral metabolic rate for

oxygen or cerebral metabolism
would be another word say that.

Structural changes in the brain.

Changes in neural melanin
one of the earliest

structures to show the lawyer accumulation
and all timers disease is the locus of.

And it's also a neural melanin rich
area so the question is whether or

not the accumulation of amyloid
might affect the neural melanin and

that can be used as
a biomarker looking at.

Gamma.

Or Gabaa concentrations which is the major

inhibitory neurotransmitter in the brain
in glutamate concentrations which can

tell us about neural inflammation
looking at white matter integrity and

functional conduct to Vittie as well
as blood brain barrier compromise these

are the different areas that we're going
to be looking at with our poll sequences.

But also in the context of
the healthy aging studies that

said we'll have blood based C.S.F.
based the biomarkers.

And the logical testing and

a lot of information about lifestyle such
as levels of self reported activity.

Diet and so forth and so
we'll have all of those things to.

Also bring to bear in looking at the
interaction of our neuroimaging biomarkers

with those other variables so
it really is quite an ambitious study and

the idea of the Healthy Aging
study was to try to have

some kind of a breakthrough impact on
how we look at also armors disease and

identifying biomarkers
Having said that you know

there probably will be other types of
disease processes we can look at as well.

OK so the.

Neuroimaging pilot study phase one which
is the phase that we're in right now.

Is looking at collecting data from.

A senior somewhere.

Well somehow I get this reversed now.

M.C.I. patients that.

You know hide in that.

Head are all timers that are.

Amyloid positive.

About thirty of those folks yes sumption
is that their amyloid positive their most

likely program all Simers disease looking
at asymptomatic older folks who are also

amyloid positive but
cognitively normal looking at a.

Symptomatic older folks who
are amyloid negative and

then looking at young folks
because as I said earlier our

idea is that we have an aging
by disease interaction here and

so we want to look at disease relevant
factors but also look at aging

relevant factors saloon look at the aging
relevant factors by comparing young to old

will look at the amyloid substrate by
comparing the cognitively normal older

folks who Ramel only positive to those
who are amyloid negative in the disease

factors my comparing older folks are
randomly positive to people with him C.I.

who are amyloid positive so it gives us
a way of breaking down this space into

an aging component a disease
substrate component and

a disease component more or less and
beginning to look at things

the neuro imaging measures which
are we're doing in two sessions

are we've got the key one awaited him or
I the T.

to flare the diffusion weighted imaging
you know from which you can do track

dog or fee and.

Measures of white man or
integrity arterial spin labeling

which measure cerebral blood flow
quantitative susceptibility man.

Being which measures iron deposits
cerebral vascular reactivity

is going to be measured by giving
people hyper Kapnek challenge by having

them inhale a higher level of C O two than
usual which will dilate the vessels and.

Give us some idea of what the maximum
bold response is which will.

Allow us to do some interesting
things to get to in a few minutes

a neural melanin scan to look
at neural melanin rich areas

particularly the look of surreally of so
that we can also do the substantia nigra.

Resting state functional
comic to video M.R.I..

Regional cerebral metabolism we can
do with a trust sequence that was

done in on John Woo's lab.

Gabba M.R.S..

Mylan water which we'll talk
about in a couple of minutes

neuro inflammation is done by looking at
glutamate concentrations is the way we're

going to do that so
that's also done with the M.R.

spectroscopy and then dynamic
susceptibility contrast which involves.

Injection of the gadolinium
based contrast.

So.

Why did we choose all these things
let's talk a little bit about that.

And you know the idea here is eventually
we'll go to phase two are going

to collect a thousand or more subjects and

what we want to do in phase one is by
doing the comparisons with the disease and

maybe even the predictors between who
has the amyloid stub straight and

not is to pick the scans
that are most promising to.

Dip to pick biomarkers for a larger sample
and that's the phase that we're in now.

So here than neuro aging
neuro imaging protocols for.

Phase one pilot study
the first is gab M.R.S.

as you all know gab is the major
inhibitory neurotransmitter in the nervous

system and in terms of negative
bold responses to task where your

bold activity actually goes down instead
of has been shown by North off and

others to be related to the gavel levels.

In the brain correlate with
the negative bold responses.

So basically N.M.R. spectroscopy is a
little bit different than those of you're

used to doing functional imaging you
pick up a large voxel in our case of

three centimeter by three centimeter
by three Samina centimeter voxel and

you basically do.

Get a spectrum of where you
can identify the levels of

different neurochemicals which include
Gabbidon as you can see here relative to

even glutamate where the signal is small
and you get the gloomy peaks in two areas.

But relative to even even
glutamate which is small gab

is smaller so
you have to have pretty good M.R.

spectroscopy images these images
are courtesy of Lisa Krishnamurthy.

So why did we choose Gabba spectroscopy
Well we've been looking inhibition

ageing for a long time but
gal published a study in.

Two thousand and thirteen in neuro
image where they showed that M.R.S.

in the media frontal cortex.

Has a point six the gavel levels have
a point six correlation with ageing

What does that mean you're explaining
nearly fifty percent of the variance.

In your Gabba signal by a person's
age no other factors and we

know we can change that arm We're pretty
sure we can change that with exercise or

we've got good indications we can so
that accounts for forty six percent of

the variance in that particular area so
it seemed like something

good to do what work of we've done in
inhibition and I'm using the royal.

We here this first slide is.

About work done by Keith McGregor some
of you know key and basically Keith uses

a measure called the silent period yes
people squeeze the Manipulate and I'm in.

Around fifty percent of maximum force and

then he stimulates the motor cortex and
in the Ipsa lateral hand of

the stimulation you get a decrease
in the E M G activity for

a few tens of milliseconds after
you know somewhere between

you know sixty to eighty milliseconds and

this is an inhibitory phenomenon that we
think that people have shown is driven or

can be driven by Gabaa be activity.

Using back off and.

As a.

Drug and.

So what happens what
Keith this showed in his

early studies is that for
send Terry older adults the it's a lot of

silent period is shorter than for
active older adults and

it shorter for active older adults
than it is for younger adults.

And he has showed that the absolute

silent period is also reduced
at midlife for sedentary.

Adults midlife.

And what do you also showed
which is unique in this study

is he also divided his young group
into active and sedentary and

there are no differences in the Ypsilon
silent period for younger adults in

the twenty to thirty five year old
age range I think is what that is.

So this is something related
to sedentary and aging.

And what Jonah.

And and Keith have shown recently
is that you can make the Ipsa

lateral silent period long again by
engaging in a row because exercise.

So you're up Celera silent period
begins to look more like young folks.

OK well that's motor stimulation
with T.M.'s basically and

evoke response and measuring
the it's a lateral silent period.

And he's done other measures with T.M.'s
that that haven't showed a difference.

What about things like language
which is something that we

have done a lot of studying our lab This
is Christina wearing a dissertation.

And one area that she noticed.

She gives a picture naming study
having folks name pictures

it's actually more complicated than that
but not worth getting into at this time.

And she showed in right parts triangular
Eris which is a part of the right

inferior frontal gyrus that
older folks showed a typical.

Bold response.

But younger folks after small
initial positive inflection showed

a large negative deflection in this area.

Which is kind of similar to what Keith
was seeing in the motor system right so.

You know this is opposite.

His was.

In the hand that was opposite to

the hand connected to the motor cortex
that was being stimulated right and

this is cortex opposite to language cortex
that should become active in picture name.

What Marcus mines or
came along in did after that what's.

Was to look at accuracy levels and to
correlate that with that bold signal and

basically what he's showing in
ageing is that if you're able to

maintain a negative bold signal.

Your level of accuracy is higher
this is the bold signal here and

if the bold signal goes below baseline
when you're doing semantic fluency

essentially I give you a category you tell
me words from the category is that you

have a higher degree of accuracy and
semantic fluency the lower your bold

signal is this occurs in the frontal lobe
but also in some post your areas and

he replicated that in two thousand and
twelve.

So one interesting thing about this.

Bold signal in the.

Write him a sphere is that
if you do aerobics exercise

you get a decrease in bold activity
during semantic fluency and

semantic fluency improves so
the idea here is that

activity in the right hemisphere
is probably not good OK so

I don't want to throw out Harold in Tire
Lee right you guys know Harold the human

spirit case in a tree reduction in older
adults is the compensatory mechanism.

But in this particular instance it appears
that the right hemisphere activities

probably actually interfering with
naming based on our study we've also

done some T.M.'s research which is
consistent with that as well and

also if you look at the regression of
V O two max on the ball change during

semantic fluency you
also see that your VO two

max is correlated with
the decrease in bold signal so

the more fit you get all
the lower the bold signal goes.

OK And so.

That is the reason why we're
doing Jabot spectroscopy because

we think that that Gabba is
driving a lot of what we're seeing

in both cognitive and
motor events in aging.

By the way also the other thing that Keith
has shown that I didn't mention is your

motor performance also improves it's not
only the it's a lateral silent period but

your motor performance.

So there's also as I mentioned
earlier a glutamate signal this.

Spectrogram has not been even doubt.

In terms of the baseline levels for
the the different areas of signal but

there's glutamate signal here and that
glutamate signal basically the higher.

When you have a neuro inflammatory process
as you're good to mate signal gets reduced

so we're going to be measuring glutamate
levels in that same voxel and.

Those the lower to me would be
an indication perhaps of neuron

Flamm atory processes which
are important in all sinners disease

so cerebral metabolic rate for oxygen.

This is a study that
comes out of Han John.

And basically what he shows that
on the average as you get older

your cerebral metabolic rate for
oxygen actually increases.

And that's seen as a little bit
paradoxical I guess in some places.

But there's the approximate
rate of being creases.

Somewhere in the one and
a half to two percent neighborhood

predicting life but
what happens when you get.

M.C.I. is that your serial metabolism
all the sudden starts to go down

in a fairly large proportion
relative to the increases

that you see with the aging So one
question that we have here is whether or

not there's a point of inflection where
you change from increasing to decreasing.

Metabolism that could be predictive and
of course you know that

scatter of points was fairly large
that I showed you earlier So

one question is what's the reliability
of your individual measure because

those were different subjects all those
data points you're just looking at and so

there's a reasonably good interest session
correlation well you're correlation

between sessions actually gets lower
as your variability goes up so

we don't know how good that measures going
to be but we're going to take a look at it

anyway I should have probably put the scan
up the way that's done is by getting your

arterial oxygenation levels
with pulse oximetry you know

on your finger but looking in your Venus
system in the brain so that gives you

an idea of what the blood going into the
brain as caring in the way of oxygen and

you take a measure in
the superior sagittal sinus of

what the oxygenation levels are with the a
mirage and that will give you an idea of

what your whole brain tabel is and looks
like since the superior sagittal sinuses

draining large areas of the brain
there are other ways to do

this get higher resolution but
that takes more time and

since we're doing thirteen sequences
times that a little bit of a premium so

we decided to use that
sequence it's called that.

The acronym trusts the Quins.

In terms of box a base more Fama tree.

I'm not sure that this measure
actually holds a lot of promise

because what happens is
where it's particularly.

Useful for predicting conversion
from M.C.I. the ADA he is

looking at your medial
temporal atrophy and this is.

Three years before.

This is one year before I can version and

I believe this is that conversion if I
remember them the images correctly but

the reason for doing this actually is that
if you can predict the beginnings of the.

Temporal atrophy with another
biomarker you can kind of ratchet back

your prediction backwards in time so
it's a way of.

You know taking something earlier and
him see I and in ratcheting

your prediction backwards and times if you
can find something that predicts this.

Decrease and temporal lobe grey matter.

Out of paying whose lab.

Jumping.

Did a poll sequence.

Chan was the first author on that
published in two thousand and

fourteen showing that they
could identify areas with.

Neural melanin.

Sensitive sequence in
the locus rule us and

other labs have showed
that if you look at.

The contrast to noise
ratio by verbal kind.

I Q As measured by
the test of adult reading

that there is a correlation between
the contrast to noise ratio for

Neural melanin older adults and
no correlation in younger adults so

the idea is that the neural
melanin sequence.

Given the fact that the locus
earliest is involved relatively or

can be involved relatively early in all
timers disease may give us an early

signal so one of the things we're going to
do then is to look at neural Mellon and

contrast and
particularly in the locus of really it's.

So in terms of diffusion
where you did imaging there.

A number of different ways you
can analyze your diffusion waited

images in basically what.

Derby and folks were doing is looking.

At whether or not they could distinguish
healthy controls from M.C.I.

that is amyloid positive.

And they looked at
different indices of both.

Structural imaging scans and
diffusion weighted imaging.

And the best index they found in
this particular prediction was

was to mean diffusive a D..

And in particularly when
the correct the mean diffusive

of the with the regression equation for
age and other factors they get

the best prediction seventy
seven percent correct and

the instinct thing about this is they
they put fraction on I suppose and

isotropy mean mean defeat.

Motive and I stop trape white matter
volumes from the structural images

grey matter volumes from the structural
images all those things plus the age and

the other factors that
they were using into a.

Support vector machine and what they.

Support vector machine and
actually the single measure

corrected with the regression
equation from mean diffusive and

he was actually a little
bit better than the.

Data from the support vector machines so
mean.

Measures might not exceed
this particular one but

in this case the multi kernel
support vector machine was not

able to produce better results
in just using a few seventy.

So that's one measure that
will probably be looking at.

And one of the things you probably know.

Is that diffusion
kurtosis imaging which is

thought to give you more
accurate measures than D.T.I..

Diffusion tensor imaging is one thing that
we're going to try on that but we've also

got measures that can look at multiple
diffusion directions in a single voxel and

and we'll be using those measures
as well with those skins.

To look at the.

This or the white hyper intensities in
the white matter that occur with the.

We're going to use a.

T two aided flare image and
automatic segmentation of those areas

of the system you can see the hyper
intensities in the three percent of Luke.

Three percent of the vine Maki pipe by
the curry osis around the ventricles in.

Greater.

Volumes of Luke where Yosef
you can see it spreading into

the corona radiate and
if you go higher into the center.

And so forth and see whether or
not those changes.

Predict cognitive changes.

We're going to be doing mile in water
a fraction the idea behind mile and

water fraction is that there
is that the environment

in which water is found affects
that its teeth to decay and.

In particular there is water
between the layers of Mylan that

surround the axons in the white matter and
that T.

to decay is much more rapid.

And then if you look at free
intra an extra cellular water and

has a much longer T two K.

T two D.

K.

and.

Basically you can look at.

Mylan water levels and
compare them to the.

All of the water in the brain
the Mylan water plus the free and

then get a mile in water fraction is
essentially what you're doing there.

And.

Colleagues basically showed that.

There are areas Mylan water content and

water fraction associated
with amyloid precursor.

Levels of amyloid precursor protein
as the blood base biomarker

I think it's blood based.

If I remember correctly.

So.

You know the idea here is that
your Mylan water fraction may

be a very sensitive indicator that
your mileage is beginning to.

Decay which is probably related
to these and so forth and

in fact what Dean and
colleagues also show is that if.

If you take your.

Measures of Fosse the town
the cerebral spinal fluid and

you divide your subjects
at the median into high and

low levels of Fausto tau is that your

subjects with higher levels of
Fausto Tao show a significant.

Reduction as you go across ages and
that's the cross-sectional study but

you can see the correlation
line between age and

decreasing Mylan water
fraction is fairly high.

With subjects who have.

High levels of Fosse with Tao are one
of the also amorous biomarkers.

Resting state F.

M.R.I. basically they're using
a measure called pair mutation entre P.

I don't really have time to explain
that now but ask me later and

I'll be happy to talk more about
what peer mutation entre P.

is.

This is probably not a way that we're
going to analyze our resting state

if I'm right data and the reason for
that is where we want to do get something

that's sensitive to the difference between
normal controls and an early M.C.I.

in this particular population
would be what would be ideal for

us and none of these particular.

Areas of the the cortex do you get that
distinction so pear mutation entre P.

is probably not a very good measure for
us.

And.

Arterial spin labeling will give
us cerebral blood flow which is

you know correlates with Pat
in Alzheimer's disease so

it's a useful substitute for
F.T.G. although.

Lisa Krishnamurthy will probably
be working on an M.R.I.

based way to do ghost metabolism.

But you can see that older persons show
decreased cerebral blood flow particularly

in the frontal cortex
the red areas are higher.

Yellow areas have somewhat lower blood
flow in the green areas are very low but

those are really actually the ventricles
of there's not much blood flow there.

But you can see the cortical ribbons the
place where you're getting high levels of

blood flow and it's higher especially in
the frontal lobes and younger than older

folks and these images are courtesy
of Lisa Krishnamurthy.

And vascular reactivity as
I said earlier basically

you're dilating the blood vessels
looking at your bold signal and

you can think of the bold
signal as being in.

In your sort of Ashley reactivity is
giving you a maximum increase for.

Your bold signal and you can and
what happens interestingly enough.

You don't see a lot of change from

an elder from elderly
controls to amnestic M.C.I.

at least in this particular study coming
out of Heinz young lose a lab but

what you do see is that
older subjects show.

Cerebral vascular reactivity than younger
folks and in fact what John John and

his colleagues did in
one study is basically

to correct your bold response in.

A.

I think it was an image viewing
task that they were using and

in in V one in the media
of Temporal Lobe basically

the uncorrected responses which are in
blue showed significantly lower.

Bold response in young than old people but
when you correct for

C.V.R. there's no difference
between young and old.

And indeed in.

Other areas of the cortex during this task
and particularly in fear frontal gyrus

when there was no or
minimal activity change for

the uncorrected signal when you correct
the signal you get much more robust

changes and there increases in
activity rather than decreases and

in this study when they corrected for
C.V.R.

they didn't find any decreases in activity
for the young relative to the older

subjects which they did find if
you don't correct the signal so

there's you know beginning to
emerge a degree of concern in

your grant reviewers and
in your article reviewers and

terms of comparing old and young with F.

M.R.I. bold activity as to whether or
not you should be correcting for

Siri revascularization
activity changes and

it's still I think a little iffy depending
on who your viewers are as to whether or

not they do that in fact I just advised
a colleague of mine not to mention

cerebral vascular reactivity because she
wasn't going to be able to correct for

it and just to send it in and
see what happens and if they ask for

it she can figure out a way to do it maybe
set out a cycle and and resubmit you know.

The worst thing you can do.

You know do a breath hold or
you can also use resting state to divide

devise an egg on an estimate of
sort of a basket reactivity.

We're doing dynamic
susceptibility contrast and

there is an interesting study.

Suggesting that there's a particular
part of the blood brain barrier that

gets compromised and
you can detect that compromised using.

Dynamic susceptibility contrast which is

you know essentially with
the gadolinium based contrast agent.

This is just a picture of
the neurovascular unit with the in Ophelia

being the kind of first line of the blood
brain barrier but it also involves.

A.

The Astro City in feet
surrounding the blood

vessel of the peri bachelor
space the basement membrane but

particularly the Perry sites the Perry
sites are what get compromised and

here you can see in the hippocampus
the brighter colors indicate more of the.

Gadolinium contrast getting into
the tissue which is higher for

older adults even higher for him C.I.

This is a multiple
sclerosis as a control and

then the young controls it's even
in aging you get some leakage

of the blood brain barrier and
you can see that happens.

In CA one ca three the dentate gyrus and

the hippocampus proper in all those areas
so that's another thing we're going

to be looking at and finally we're
using quantitative susceptibility man

being and you can do a couple of
things with that one is you can.

Locate microbe leads which may be related
to him ridges Will why is that important

you tend to get vascular calm
compromised there's amyloid with

the in the areas of amyloid deposition and
you get tend to get micro bleeds around.

There so you can detect some
microbe leads with that.

Quantitative susceptibility mapping which
is taking advantage of the fact that

iron deposits in the brain from the.

Blood in the micro him EJA
him ridges will show up but

you can also exclude those
micro him ridges and

just look at Iron deposition and
this is the one thing that I think is very

prescient because what we're looking
at here is different cognitive tests.

And these are the domains
that tests come from.

And you can look over years period
of time at the change in the.

Score of these particular
cognitive tests right so

and this is standard
deviations per year and

what you will notice here is if you
look at the amyloid positive which

is the yellow line and
you know Malloy negative that.

In the amyloid positive
subjects what happens is your.

Your level of.

Iron deposition in the tissue
predicts cognitive decline so

it's an interaction between
your cognitive function and

your iron deposition and
your amyloid status here and

this is the one I think is
particularly present prescient for

those of us studying cognitive ageing
because what it shows is this interaction

between oxidative stress is what the iron
in the tissue is thought to indicate

am a lawyer and and
that becomes important.

Only for the amyloid positive
group in terms of predicting

decline in cognition and so
these are some of the thanks.

It gives you an idea of I think
where we're heading in the future.

In terms of studying cognition and
aging and

I've gone a little bit over mine a lot
of time but you can kind of get an idea

of where this project is going and
maybe some ideas

about what's going to be going on in
studies in aging and in the futures.

Thanks for your attention.

We have time for questions of Cripps.

I'm still looking at the food.

Feel free if you have questions.

I know it's two visits
about usually two and

a half hours we have some
subjects if they live far away

we're willing to image them in one day but
we don't like doing

that because you know the more you lie
down the more restless you get and

you know the higher the pain from
the head corals the kind of tight and

so forth and you start to get head
motion towards the end of an hour and

a half worth of imaging and.

Even if you take somebody out for hard
put it back in they're still restless and

it kind of compromises the imaging So you
like to do the first session of an hour

and a half in the second session of an
hour in the last thing we do in the second

session is the dynamic susceptibility
contrast because the gadolinium is going

to hang around for a while and you know
affect all of our other imaging So

it's the last thing in the protocol.

No this is.

We're in a we're in phase one right
now where we're still looking

you know we're going to eventually not
do the young and not do the amyloid.

Positive M.C.I. folks and just concentrate
on the cognitively normal older folks and

look and see whether or not we can use
our biomarkers to predict the difference

between the amyloid positive
in the amyloid negative seeing

the amyloid positive theoretically is
being asymptomatic all timers disease or.

I guess we wouldn't quite call it progress
that's really more the term reserved for

him c I But you know that's a thought so
when we go to the Thousand are only going

to be looking at older folks and
looking at the diversity and the positive.

Yeah right that's the Right now we're
there and the reason that we're doing

that again is we want to be able to
pick the most sensitive biomarkers

to study the Healthy Aging and
then you know the idea is that

at some point in time in the not
too distant future we're going to

support this by grants from an I.H.
Once we've established the cohort.

And so forth.

And we should have launched today the
whole data and all of the biomarkers that

we're looking at as Wells connoted testing
although as you can see from these.

Diagrams you know when you repeat
cognitive testing sometimes

there's an advantage.

To you taking the cognitive tests again

because you can see in
the amyloid negative folks.

Yeah yeah yeah yeah yeah.

Yes Yeah.

Stream very very.

Well there's a.

Three years of research with the C.S.F.
biomarkers they've established a level

of C.S.F. biomarker that
they feel is significant and

oddly enough in the amyloid you'd think
it would be more C S F amyloid but

it's actually less is a bad
thing in terms of the C.S.F..

But they've established
a threshold that they feel.

Even clinically these days
a lot of people that go to.

Places that have eighty or
Cesaire using C.S.F.

amyloid cutoffs to help them
understand cognitive changes and

complaints and older people so
it's a cut off Korea.

Correct yeah yeah that's true and

in fact you know what might be
the continuum is probably going to be of

more predictive value than splitting
them but you know for the sake of.

Simplifying their analyses.

They divided into namely positive and

when you do you know Chris So
I mean one of the interesting.

Studies showing.

It's.

Five.

Hours project come on.

You're well yeah right
it's just you know so.

Yes.

Well probably choose the most promising.

Indices from each image type.

You know that we can figure out
based on the literature and

put them into an ass V.M. with other
biomarkers and look at whether or

not you know what kind
of prediction we get but

we'll also be looking at the images
individually you know to understand them

on our universe at level if you
will in an individual image level.

You know that's not a political
law like we get one value for

the gabble levels and
you know one level of value for.

Cerebral metabolic rate for
oxygen but another you know like the.

Images where we're analyzing them in a
voxel wise level yeah we'll be looking for

areas and probably doing exploratory
analyses or and or well hypothesis driven.

Region of interest level.

So yeah well we'll probably doing all
those things hopefully we've got enough

analysis horsepower with three or
four people working for

us that we should be able to
both feed data into an ass V.M.

where we can put all the other viral
markers in it and you know use

multivariate algorithms to try to get
the best detection combination but

we'll be trying to understand regional
differences and individual image level and

smaller level where that more targeted in
an interactions as well I mean there's

a lot of things we can do with these data
a lot of ways that can be analyzed and

be thirteen image sequences but also you
start thinking you know probably I don't

know we've got probably fifty sixty one
hundred other biomarkers genetic markers.

Cognitive functions you know so
I you know I think

the real issue is to think about what
it is we're trying to predict or

what it is that we want to find out and
then tailoring the analysis to do that.

I think that's going to be the real
challenge is thinking this through

in a way.

That we don't capitalize on chance and
do the right.

Target the right.

Indices.

For that.

But.

Yeah I mean that's a really important and
yes I would assume

with that many variables there's going
to be plenty of multi colinear T.

that we need to avoid in
terms of how we analyze and

model the data yeah I think
that's a very good observation

and your questions well I was enjoying
coming out and talking to you guys.

Happy to take any other questions
an individual level if you want.

I don't know.

I don't know that we've decided that I
think the model that we're currently using

is more you know people proposing studies.

And looking at those are we going
to make them publicly available.

I don't know I think that's a good
model and on the long run I think there

are some biases toward that I don't know
that that decision's been made perhaps

it has I don't know I assumed you know
we're going to want our investigators to

to get a first crack at them but
I don't know that you know

if somebody were interested in
a particular set of biomarkers or

something like that or had a particular
question and if they came to us and

wanted to look at it you know we go wow
that's an interesting idea let's talk

about how we can do that together like
that I mean I'm open to that I can

tell you that and there are more data than

you know we're going to be able
to think of every analysis for.

You know from my standpoint if
somebody were to come to me and

do something that makes a lot of her
want to do something with the data that

make a lot of sense I would
say OK let's talk to.

The gunas Yeah yeah yeah absolutely.

The students it's something.

Yeah one more.

Thing.