►
From YouTube: ACTION-EHR: Patient-Centric Blockchain-Based Healthcare Data Management for... Alevtina Dubovitskaya
Description
ACTION-EHR: Patient-Centric Blockchain-Based Healthcare Data Management for Cancer Care - Alevtina Dubovitskaya, HSLU (Lucerne University of Applied Sciences)
Due to the privacy and high sensitivity of Electronic Health Records (EHR) and the lack of systematic infrastructure support, most EHR data sharing still happens through fax or mails, which can incur major and possibly critical delays of patient care. Therefore, there is an impeding need for a patient-centric system that will facilitate secure and trustable management of EHR data originated from multiple hospitals, which is often the case, in particular, for patients with chronic diseases and serious medical conditions such as cancer.
Researchers from the universities in US and Switzerland, and in collaboration with medical doctors from University Hospital in US, employed Hyperledger Fabric to build ACTION-EHR - a prototype of patient-centric blockchain-based EHR data management system. A Web application is developed for patients and doctors for initiating EHR sharing transactions. A hybrid data management approach was chosen. The system ensures patient privacy protection and security guarantees with respect to the requirements to the healthcare data management, including the access control policy specified by the patients.
Alevtina will present a demo of the prototype that has been tested in a distributed environment at Stony Brook University and demonstrated the feasibility of applying Hyperledger Fabric to address various inefficiencies of health care-related processes, by enabling better traceability, transparency, and efficiency. She will also discuss major challenges of mass adoption of such system. In particular, she will focus on current capabilities of Hyperledger Fabric from technical, legal, and social perspectives when developing blockchain-based system in healthcare settings.
----
Hyperledger – Open Source Blockchain Technologies
Hyperledger is an open source community focused on developing a suite of stable frameworks, tools and libraries for enterprise-grade blockchain deployments.
It serves as a neutral home for various distributed ledger frameworks including Hyperledger Besu, Fabric, Sawtooth, Indy, as well as tools like Hyperledger Avalon, Cactus and libraries like Hyperledger Aries, Ursa. Learn more about Hyperledger projects: https://www.hyperledger.org/use
Case Studies: https://www.hyperledger.org/learn/case-studies
Training & Certification: https://www.hyperledger.org/learn/training
Tutorials: https://www.hyperledger.org/use/tutorials
Webinars: https://www.hyperledger.org/learn/webinars
Events: https://www.hyperledger.org/events
Vendor Directory: https://www.hyperledger.org/use/vendor-directory
Subscribe to the Hyperledger Newsletter: https://www.hyperledger.org/newsletter
Follow-us on Twitter @Hyperledger
Learn about Hyperledger Membership: https://www.hyperledger.org/about/join
#Hyperledger #HyperledgerGlobalForum2020 #Blockchain
A
Hi
everyone,
my
name,
is
Elif
Tina
and
I
would
like
to
start
by
dragging
your
attention
to
this
health
data
map
that
was
created
in
Harvard
University
some
years
ago.
That
shows
how
many
different
entities
process
and
exchange
our
health
care
data
in
anonymized
and
long
anonymized
fashion.
So
this
with
this
number
of
interactions,
it's
actually
it's
impossible
to
keep
track
of
where
our
data
are
and
who
can
access
it
and
who
is
processing
it
right
now.
A
But
actually
it
would
be
really
nice
to
know
where
our
data
are
and
be
able
to
exchange
it,
because
this
can
improve
personal
treatment
outcome.
It
can
improve
medical
research,
create
possibilities
to
conduct
a
more
clinical
trials
and
even
bring
the
health
care
to
the
new
level
like
use
the
data
that
we
all
produce
in
order
to
enable
AI
driven
health
care.
A
For
this
type
of
violation
of
privacy
and
the
government
actually
support
this
initiatives
to
allow
the,
inter
probability
between
different
different
healthcare
databases
or
different
healthcare
data
sources,
but
ensuring
that
the
patient
can
access
the
data
and
can
decide
with
whom
he
wants
to
share
what
kind
of
data
for.
But
for
what
time?
For
how
long?
A
Sometimes
people
are
really
concerned
about
their
privacy
and
they
don't
want
to
share
the
data
for
research,
but
there
are
also
other
peep
that
would
like
to
share
the
data
as
they
are,
but
because
the
data
by
default
is
being
anonymized
for
research,
then
we'll
lose
a
lot
of
utility
of
the
data.
So
there
are
different
approaches
that
tried
to
address
this
challenge
like
centralized
approach
or
using
the
health
information
exchange.
A
However,
we're
still
asking
ourselves
how
to
ensure
interoperability
and
privacy
simultaneously,
and
this
brings
us
to
the
question-
why
would
we
use
blockchain
for
that?
Not
for
the
sake
of
using
of
blockchain
and
I'll?
Try
to
convince
you,
so
the
idea
of
so
we
started
working
on
that
on
that
idea
some
years
ago,
and
the
initiative
actually
came
from
the
medical
doctor
that
knowing
the
some
basic
principles
of
the
distributed
letter
technology
notice,
that
there
are
a
lot
of
similarities
in
the
sense
of
this
electronic
health
record
management,
environment
and
the
blockchain
technology.
A
So
especially,
for
example,
there
is
this,
distributed
data
sharing
architecture
where
peers
have
to
need
to
have
the
the
same
data
you
need
to
process
the
same
data
or
where
we
need
to
keep
some
immutable
history.
So
this
is
the
same.
We
need
to
keep
the
we
need
to
keep
the
old
history
of
access
to
the
data,
especially
for
the
court
cases
in
healthcare.
So
the
also
the
permission
blockchain
technology
allows
us
to
develop
the
efficient
system
that
would
allow
this
efficient
and
fast
data
sharing,
which
is
sometimes
really
life
over
life
importance.
A
So,
of
course,
we're
not
going
to
put
healthcare
data
on
the
blockchain,
because
it's
very
sensitive
and
because
of
the
volume
of
such
data,
but
what
we're
going
to
put
on
a
block
change
the
system
metadata.
That
includes
the
links
to
the
references
to
the
healthcare
data
that
is
stored
securely
and
encrypted
on
the
cloud
server
and
permissions
permissions
specified
by
the
patients
for
all
these
different
categories
of
file
or
even
the
files.
A
So
we
we
actually
had
our
first
prototype
the
the
long
time
ago
of
our
fabric
of
with
fabric
0.6,
where
we
modeled
our
blockchain
network
as
having
different
hospitals.
So
even
doctors
coming
together
and
then
running
the
chain
code
that
with
just
the
very
basic
operations,
just
to
add
the
permissions
or
add
metadata
information.
While
adding
the
data
on
the
cloud,
then
we
would
have
the
membership
service
that
would
verify
the
entity
of
every
doctor
every
hospital
and
then
the
solution.
A
But
of
course,
so
the
problem
of
such
system
is
that,
of
course,
it's
not
it's
not
possible
to
scale
too
much,
and
even
if
we
think
that
we
can
maybe
use
this
validating,
no
there's
the
very
big
hospitals
and
then
then
be
still
with
the
evolution
of
hypervisor
fabric.
We
also
redesigned
our
system
architecture
where
currently
we
have
so
we
redeployed
it
on
1.0
than
one-point-two,
and
now
we
have
it
on
now,
1.4,
which
takes
actually
a
lot
of
students
and
and
master
projects
to
always
keep
up
with
their
changes.
A
But
so
currently
we
have
we
model
every
hospital
as
a
hyper
laser
fabric
organization
and,
basically
to
start
the
network.
You
just
need
to
have
one
admin
hospital
that
would
admin
organization
order
and
the
cloud
storage.
So
then,
the
other
hospitals
which
are
represented
as
a
hyper
ledger
of
fabric
organizations,
can
join
the
channel
and
in
every
so
this
allows
us
to
have
in
each
hospital
to
have
its
own
role
of
administrate.
A
That
can
actually
manage
the
identities
of
the
doctors
and
patients
and
then
also
we
edit
the
possibility
not
to
just
to
share
the
data
and
upload
the
new
permissions
based
on
which
the
data
from
the
cloud
will
go
then
shared
with
the
doctors,
but
also
to
have
the
possibility
to
comment
the
files
that
that
have
been
seen
by
the
doctor's
command.
The
healthcare
data.
So
here
is
just
a
few
screenshots
from
the
from
the
logs
of
so
I'm
talking
now
about
creating
exactly
the
same
system.
A
The
system
is
shown
here,
so
I
will
have
first
admin
hospital,
and
then
we
have
just
another
hospital
joining
the
network.
So
we
have
two
organizations
and
we
basically
start
with
the
just
building
with
the
having
two
peers
of
the
first
admin
hospital
joining
the
channel.
And
then
we
start
our
web
server.
Yeah
I
should
have
said,
maybe
here
that
we
also
in
each
hospital.
We
assume
that
there
is,
we
will
build
the
web
server.
A
That
has
this
all
the
local
health
care
data,
plus
all
the
users
registered,
and
we
also
have
a
web
server
where
the
registered
patients
and
doctors
can
use
in
order
to
query
and
update
the
ledger.
So
here
we
start
by
just
adding
two
peers
to
the
channels.
Also
we
having
just
one
organization,
then
we
create
the
config
file
of
the
hospital
too.
That
wants
to
join
this.
This
network,
this
new
organization
and
we
install
the
chain
code
and
then
send
the
information
about
the
order
to
the
new
joining
or
beer
from
the
new
organization.
A
Then
now
you
see
the
screenshot
from
this
new
organization
where
we
see
how
the
peers
from
the
new
hospital
are
joining
the
chain
code,
and
then
we
see
that
the
last
blog
that
was
created
with
the
latest
configuration
file
is
retrieved.
Now
on
the
on
the
second
hospital
that
is
joining
the
network.
So
then
we
have
the.
A
We
have
the
fear
from
the
new
joining
hospital
joining
the
the
channel,
and
then
we
have
the
sending
the
information,
and
then
they
appear
from
the
initial
from
the
anchor
Hospital
sense.
The
configuration
learns.
The
configuration
of
the
new
organization,
then
then
we
also
need
to
upgrade
the
chain
code
and
we
have
the
network
of
two
organizations
order
and
a
cloud
server.
So
in
the
demo
that
I
hope
I
still
can
show
you.
A
We
have
now
so
now
we're
moving
from
from
the
back
end
to
the
the
web
interface
and
to
see
what
actually
now
we
can
do
with
with
this.
So
we
have
the
network
of
two
hospitals
organization
set
up
and
with
the
patients
and
doctors
enrolled.
We
want
to
demonstrate
how
patient
from
the
first
hospital
can
share
the
data
that
is
currently
reside
on
the
in
the
hospital
one
with
the
doctor
from
another
hospital
that
just
joined
our
blocking
Network.
A
We,
while
this
is
happening,
the
data
is
also
being
encrypted
with
the
public
key
of
the
doctor,
with
whom
the
patient
wants
to
share
the
data.
The
data
is
being
sent
to
the
cloud,
and
then
this
sharing
transaction
is
reflected
on
the
ledger,
based
on
the
take
into
account
for
the
the
time
for
how
long
the
patients
went
away.
She
wants
to
share
the
data
and
then
we
demonstrate
how
the
doctor
can
see
and
can
access
the
data
and
also
comment
and
see
other
comments.
A
So,
first
of
all,
as
I
mentioned,
there
is
this
admin
role.
So
here
now
we
are
on
the
first
Hospital,
the
initial
Hospital
that
was
created,
admin
hospital
and
we
can
see
basically
the
list
of
the
doctors.
We
can
also
add
new
doctors,
and
we
can
also
see
the
list
of
the
patients
and
also
through
this
admin
console.
We
can
also
add
new
patients
to
the
enroll
new
patients
in
the
hospital.
Now
we
are
going
to
use
an
example,
this
patient,
in
order
to
use
now
the
patient
interface
in
order
to
actually
share
the
data.
A
So
now,
when
the
patient
logs
in
we
can
see,
you
can
see
his
permissions
and
he
can
also
see
yeah
permissions
given
to
different
doctors
and
also
list
of
caregivers.
So
now
we're
going
to
share
the
data
category,
which
is
history
and
physical
exams
with
the
doctor
from
the
hospital,
and
we
just
choosing
the
the
for
the
first
doctor
I.
Think,
because
here
we
already
see
the
list
of
the
doctors
that
are
registered
with
the
the
the
hospital
that
just
joined
the
network,
and
then
we
also
have
to
specify
what
was
a
date.
A
What
is
the
right
so
to
read
access
the
data
and
specify
the
timeframe?
So
now
we
see
that
the
data
we
don't
see
how
the
date
is
being
applauded,
but
we
see
that
the
transaction
successful
and
we
now
can
see
that
the
permission
appears
in
the
list
of
permissions
and
in
the
list
of
caregivers
that
can
access
this
category
of
the
data
about
this
patient.
A
During
the
specified
time,
interval
also
can
help.
The
patient
also
has
a
possibility
to
generate
the
report
and
now
I'm
going
to
demonstrate
how
the
doctor
from
the
hospital
that
just
joined
the
network
and
and
got
the
permission
from
the
patient
to
access
the
healthcare
data,
how
the
doctor
logs
in
and
then
this
information
that
can
be
customized
and
here
in
the
list
of
the
patients
that
shared
the
data.
We
can
see
that
there
is
now
our
category
of
the
data
for
of
the
specific
patient
with
the
permissions.
So
the
doctor
can
download
the
data.
A
So
now
it's
being
downloaded
and
decrypted
from
the
cloud
cloud
server,
and
so
here,
where
we
currently
use,
as
I
mentioned
the
public
key
of
the
doctor.
But
we
had
we.
We
were
really
sedating
if
we
should
come
up
with
the
with
different
approach
and
use
for
examples,
patient
specific
keys
so
that
we
don't
have
to
replicate
if
the
patient
wants
to
share
the
data
with
multiple
doctors,
but
it
all
has
its
different
pros
and
cons.
A
A
Why
we're
not
there
yet
so
we
you
can't
find
yet,
unfortunately,
this
kind
of
things
in
your
hospital-
and
the
reason
is
that,
while
we,
we
kind
of
able
to
provide
the
mutability
of
all
these
permissions,
our
consents
and
we
can
encrypt
the
data
and
and
and
say
that,
we're
really
providing
this
secure
entrust
of
data
management.
However,
there
are
still
a
lot
of
non-related
blockchain
issues
that
really
not
allow
us
so
to
go
there
like
all
this
key
management
and
all
the
credentials
management.
A
Also
Identity
Management,
actually
now
we're
thinking
to
use
deeds
for
also
with
with
this
approach,
instead
of
using
the
standard
identity
management
solution.
Awesome
with
this,
what
was
one
of
the
topics
during
the
keynote
is
that
interoperability
is
the
key,
and,
given
that
there's
a
lot
of
approaches,
a
lot
of
attempts
to
to
create
this
kind
of
systems,
we
risk
turned
up
with
having
just
multiple
blockchain
networks
that
are
not
able
to
interact
with
each
other
with
each
other
and
this
this
what
makes
data
actually
this?
A
What
makes
the
availability
of
the
data
limited,
also
by
letting
patient,
to
decide
what
to
do
with
his
healthcare
data,
we're
risking
the
adapt
in
the
situation
where
patients
just
doesn't
share
his
data,
or
he
doesn't
share
the
very
important
data
that
maybe
is
really
crucial
for
this
particular
diagnosis.
So
before
before,
allowing
patient
to
check
to
really
decide
what
he
wants
to
do
like.
A
We
have
to
come
up
with
some
mechanisms
that
would
maybe
define
the
minimum
data
that
has
to
be
actually
have
to
be
accessed
accessible
in
case
of
emergency,
for
example,
all
for
some
for
some
basic
diagnosis,
also,
even
that
we're
actually
trying
to
make
sure
that
the
this
access
control
policies,
these
permissions
seem
I
exactly
the
same
at
every
joining
organizations.
Every
node
we
cannot
actually
guarantee
that
some
nodes
are
not
behind
and
we
there
is
it's
it's
really.
A
It's
really
question:
if
we
have
to
just
query
one
node
or
we
have
to
query
all
the
nodes,
each
time
we're
taking
decision
about
whether
the
data
has
to
have
to
be
shared
or
not
also,
there
are
different,
limited
capabilities
of
convolution
technologists,
even
though
now
with
moving
to
one
point
X
and
2.0,
there
is
a
big
progress
in
the
in
in
scalability
of
of
hydrologic
fabric.
There's
still
a
lot
of
there.
A
Still
it's
hard
to
imagine
that,
once
we
have
all
our
data
on,
we
have
that
we
have
a
platform
that
can
support
the
healthcare
data
or
even
permissions
of
healthcare,
that
of
all
the
patients
in
the
world.
Also,
even
we
try
to
move
to
having
the
chain
code
responsible
for
a
lot
of
things,
but
at
the
same,
at
the
same
time,
there
still
has
to
be
someone
that
is,
writing
this
chain
got
writing
this
mark
on
and
by
trying
to
remove
the
bias
and
making
sure
that
everything
is
trustworthy
and
correct.
A
There
still
has
to
be
person
behind
that.
That's
the
chicken
and
egg
problem,
I,
guess
and-
and
the
last
point
I
guess-
is
that
they're
especially
very
relevant
to
the
genomic
data.
Even
though
we
don't,
we
are
not
well
I.
Don't
think
that
anyone
wants
to
put
the
whole
sensitive
healthcare
data
on
the
blockchain,
but
yet
we
because
even
putting
just
the
permissions,
there's
already
a
lot
of
risk
evaluation
of
the
privacy,
because,
if
I,
if
I
share
my
data
with
oncologist.
Well,
this
the
speaks
by
itself
and
also
when
we
talk
about
genomic
data.
A
We
are
by
sharing
this
kind
of
data
or
giving
permissions.
We
are
also
giving
the
permission.
We
also
risk
read
the
privacy
of
our
children
of
our
relatives,
which
also
I
guess
brings
a
lot
of
questions
to
the
table
and
on
that
very
mix.
The
summary
with
so
I
hope,
I
I,
convinced
you
that
we
we
try
to
use
blockchain,
not
for
the
sake
of
blockchain,
but
there
is
still
a
lot
of
stuff
that
that
has
to
be
done
and
that
I
hope
that
all
together,
we
can
actually
do
and
all
they
say.
A
A
A
A
A
So
this
is
something
that
of
course,
can
be
so
for
now
we
just
have
doctors
and
and
patients,
but
of
course,
when
we
were
just
thinking
about
whether
it
makes
sense
to
use
blockchain
and
I
also
know
that
there
are
different
pilots
that
also
include
the
insurance
companies
and
doctors
in
this
communication
for
the
reimbursement,
for
example.
So
we
for
that
particular
project.
A
We
did
not
have
any
other
users
yet,
but
of
course,
in
that
very
optimistic
slide
once
so
in
the
hospital
can
get
all
different
kind
of
users
and
actors
that
that
can
in
the
organization.
So
maybe
it
can
also
be
the
insurance
company
that
is
joining
as
well
and
then
by
using
different
channels
having
just
access
to
subsets
of
the
information,
and
then
this
would
allow
us
actually
to
to
divide
which
kind
of
information
has
to
be
have
to
be
accessed
by
which
entity.
A
So
the
the
screenshots
that
we've
seen
and
the
the
videos
was
just
done
on
the
two
students
laptop
sitting
in
the
same
room,
but
also
we
have
but
on
two
different
machines
so
and
then-
and
we
also
have
the
installation
with
the
Amazon
Cloud.
So
we
have
one,
though
we
have
just
to
Amazon
machines,
so
we
have
like
three
hospitals
and
also
one
AWS
for
the
for
the
cloud
server.
So
we
also
got
interest
from
other
hospitals,
so
we
work
with
the
Stony
Brook
University
Hospital
and
with
that's
where
we
get
all
this
healthcare
insights.
A
A
A
But
this
makes
a
lot
of
sense
and
what
we
would
like
to
do
now
is
to
have
so
based
on
all
these
permissions
and
the
comments
of
the
doctors
we,
where
actually
it
so
now
by
having
the
ledger
that
stores
the
permissions
of
who
can
see
what
kind
of
data.
Now
for
every
doctor,
we
can
generate
the
patient
history
that
is
allowed
for
this
doctor.
So
that's
what
we
would
like
to
do
next,
but
by
processing
the
the
information
on
the
ledger
and
trying
to
create
this
kind
of
full
patient
history.
A
The
patient
because
well,
of
course,
the
data
always
stays
at
the
hospital
where
the
data
was
created,
but
when
we
say
that
well--that's
is
something
that
I
could
have
also
covered
a
bit
more.
So
when
the
patient
uploads,
when
the
patient
was
to
share
the
data
and
allowed
another
unit,
allows
the
doctor
to
access
the
data
we
set
up
the
time
for
the
the
timeframe.
What
this
permission
is
valid,
so
we
still
have
to
put
some
trust
somewhere.
A
But
this
is
still
this
there's
still
a
lot
of
assumptions,
but
I
see
the
point
of
your
question
and
the
tricky
parts
that
if
you
go
out
of
the
blockchain
world,
then
how
do
you
make?
How
do
you
ensure
that
your
data
was
deleted?
So
we
can
say
that
maybe
there
is
another
script
or
another
chain
code
that
runs
and
then
periodically
verifies
the
letter
and
then
deletes
the
data.
But
then
again
nobody
can
prove
you
that
the
date
was
deleted,
but
at
the
same
time,
what
we
also
do
is
that
we
encrypt
the
data.
A
A
So,
as
I
mentioned,
we
we
didn't
run
it
in
the
hospital,
so
we
run
it
as
the
every
organization
was
run
at
the
laptop,
and
we
worked
with
the
hospital
just
to
get
the
insights
and
to
understand
how
this
healthcare
process
work.
But
we
are
aware
of
these
challenges
and
we're
working
on
using
the
device
standard.
So,
for
example,
we
can
try
to
have
well
all
the
data
that
is
being
imported.
So
there
are,
it
can
be
still
well.
A
It
can
be
in
the
best
case
scenario
in
the
fire
standard,
as
it
has
to
be
every
as
it
should
be
everywhere,
but
also
there
are
different.
What
we
try
to
use.
For
example,
there
are
the
different
software
that
also
allow
the
conversion,
for
example,
from
fire
to
different
standards
and
then
to
use
that,
but
well
as
I
mentioned,
we
are
aware
of
this.
We
haven't,
we
are
hoping
to
soon
tested
between
the
different
hospitals,
but
we
didn't
do
it
yet.