JOSEPH COHEN: All right,
how’s everybody doing today? Participation will be
rewarded with Seattle pins, kiddoEMR Seattle pins we went. They’re Rocky Mountain pins,
stainless steel, and cloisonne. Nothing but the best. Unless you have a nickel
allergy, then don’t sue me. Just saying. There’s nothing left, people. There’s nothing left. It’s like bleeding a
stone, not going to happen. My name’s Dr. Joe,
Dr. Joe Cohen, board certified pediatrician. And I’m also founder
and former CEO, now Chief Medical
Officer, of kiddoEMR and Unificare Limited in
Austin, Texas and London, England respectively. Today we’re going to talk about
developing apps for health care. We’re not going to be talking
about how to program them, but we’re going to talk
about the Med Dev zone and how we are going to be
lending our expertise to you guys to help hack
health care, to build hashtag apps with impact. I’m a board certified
pediatrician. I finished med school
in 1999 when I was 14. And did my residency
in general pediatrics at Mount Sinai University
in New York City. And then I moved
to Austin, Texas and founded Cedar
Park Pediatrics. We’re now 10 years old with a
history of 5,000 patient charts and over 3,500 active patients. We were voted the
best pediatrician in Cedar Park, Texas which
is a suburb north of Austin. And we’re very honored
to be able to be parts of so many people’s lives. I’m also the founder and Chief
Medical Officer, as I said, of kiddoEMR. I’ve also been elected to
the advisory board of IDEAA, where I will help
developers engage health care with technology in
credible and meaningful ways. The last thing I’m
here to do is lecture. I really like discussions. So I encourage
people to disrupt me. I’ll just throw a pin at you. It’s as simple as that. So please raise your hand,
stand up, say you’re wrong. We can debate it, go
outside and fight. It’ll be fine. All right? Good. So what is our mission? Why am I here? Why am I not seeing
my patients today? My mission is to harness
the power of Android to develop meaningful solutions
that are credible and compliant with the law at scale. One of the things that’s
made kiddoEMR so successful is architecture. And we’re able to produce
solutions for scale, not just for one people,
not just for people that can afford iPads, not
just for the people that have app store accounts. We are a global thin
client, cloud based, electronic medical record
platform for pediatricians, parents, and institutions. We do not market
directly to children. And if you’re going to
make apps for health care, you may want to
keep that in mind. They’re not your audience. Your audience are their parents. You really are
opening a bag of worms if you’re not building
a game for children. If you’re building some
sort of health care app, it’s going to become a major
hurdle to get it approved and released and not sued. So what’s the problem? The problem is that
in the United States, health care spending
is out of control. This statistic is
very well known. We spend $2.9 trillion in
health care every year. That’s more per capita than
any nation in the world. It’s also more than any
nation in the world. The problem is that only
1.9 trillion of that actually goes to your health. And $1 trillion is thrown
in the garbage in waste, inefficiencies, poor practicing,
duplication of services, omission of services,
readmission, noncompliance. Anyone been to the
doctor recently? OK, so how much
of the information did you take home
from the visit? 0 or three? OK, it’s funny you
should say that. There you go. Sorry, I’m a bad arm. I went to med school, people. When everyone else
was making money off Dell in the stock
market, I was learning about anatomy and physiology. In the long run, I won. Thank you. Welcome, guys, welcome. What Mark brought
up and what I’m trying to make the point is,
at the end of your visit, what does the doctor
spend time doing with you? He educates you about why you
came here in the first place. Statistics show that no matter
how much the pediatrician talks to the parents, no matter how
much time he takes to explain, you take at the max three
information points home. I believe this is one of
the fundamental reasons why we take $1 trillion and
throw it in the garbage. Because I see patients all the
time come back to me saying, we did what you said
and he’s not better. And they didn’t do what I said. They did what they
thought they heard me say. And because there’s
no credible technology to transfer information
from a doctor to a patient, we’re missing follow ups. We’re missing follow
through and we’re missing patient compliance. Every mistake made
in health care, whether it’s intentional or not,
costs us money in the long run. If you’re a diabetic and you’re
being treated for ketoacidosis, which is a very bad
thing, you basically spend a week and a half in
ICU, three days in step down, and five minutes at discharge. Why aren’t there platforms
where information can be transferred
to the patient where the patient can engage their
care provider after discharge? Well, this is the result.
$1 trillion into the trash. $1 trillion because of poor
health information management. $1 trillion due to poor
litigation and liabilities. All right, now let’s move on. Until recently, as a
physician, a physician who’s very engaged in
the tech community, I had no medical
apps on this phone. None. None of them are worth my time. Recently found one. It’s called Figure 1. I didn’t develop it. It’s an app for doctors. It’s like Instagram for doctors. We get to show you all of our
patient rashes, and everyone comments. It’s in a HIPAA
compliant fashion, but it’s the first
engaging platform for me. And there’s just very
few clinically relevant applications that
exist and that’s kind of why where we
live and why I started. So what’s the solution? Some of you may have seen me
last year at Big Android BBQ. And the community
really inspired me to come up with
a platform to stand on where we can crowd source
with physician involvement that encourages innovative thinking
and also helps us disseminate our innovative platform. And we’re going to go into
what our platform does and what you guys will
be able to do with it as you develop solutions
for health care. The flop sweat in
here is pretty– I feel like Kathy Griffin. So what’s the purpose is
that we want, like I said, we want to engage the community. We want to make and make
clear very discreet parameters and protocols. And also provide you with
a robust test environment of granular anonymized data in
which to enforce information transfer, follow up,
communications, ratings, diagnostics, evaluation. And we’re going to add
care value to the platform and to the process of medicine. Any questions so far? These are some of the
hashtags we’ve been using. Apps with impact. It all goes back to where’s my
health care app with impact? Where’s my Facebook for doctors? Where’s my end to end
encrypted chat app so that I can talk
to my patients without violating federal law? They don’t exist. And if they do, they’re
built by large corporations and they are convoluted. They’re not side by side and
they cost thousands of dollars a month. Apparently people still
think we make a lot of money. Yes, we make money and
we do it consistently, unless you do something bad and
are put out of your profession. But the fact is that
in today’s day and age, we haven’t gotten pay
raises in decades. But that’s not why I’m here. Why I’m here is because we’re
going to disrupt health care. We’re going to take it back
from the large corporations who are non-physicians
who are trying to develop for health care. And they’re making our industry
more convoluted and less efficient. And that hurts who? Does it hurt me? Yeah, I complain about
it for five minutes at the end of the day. But ultimately who does it hurt? All these children here. They’re the ones that it hurts. They’re the ones that
gets suboptimal care. They’re the ones that have to
wait so long during the visit to see their doctor that
by the time the end comes, where the real
meat of the matter is, the reason you’re there in
the first place, it’s like, OK, I’ll see you later. That’s unacceptable. We can’t do it and
we’re not sustainable. So why kiddoEMR? With kiddoEMR, especially
in the last year, we have established two
outlets for your development– the United States and
the international market. Unificare Limited, our exclusive
reseller in London, England, is going to adjust UK and India
and eventually beyond where your technology
can be incorporated into what we’re trying to do. And we’ll also be
giving you guidelines on how to address unique
demographic populations. Remember that with
a genetic pool comes differences
in medicine, comes differences in
treatment and outcome. One of the very simple
examples is vaccinations. The United States
vaccination program is a little bit different than
the UK, but really different from India. Because of our architecture,
we can adapt the protocol APIs for that, adapt the
scheduling for that. But we really believe in an
open architecture, open source. I explain it this way. Our technology is not the
Electronic Medical Record. As a matter of fact,
our EMR is open source. The technology is what
we do with the data. Think about this. When everyone was using
a quill and parchment, that was technology. The pen and paper came
out and it was technology. But today it’s just
a pen and paper. And that’s where we’re going. Eventually EMR and the
data entry platform is really not going
to be the technology. But what you do with the
data, how you incorporate it, how you stream it, and
how you integrate it with patients and doctors’
lives, that’s our technology. When we’re using open APIs
and an agile architecture in which to address very
customized populations in a very customizable way. By the spring 2016 we will
have an anonymized test data pool, a test pool of data
where your applications will be able to use
these demographics and medical indices and values
in which to sandbox your apps and play with them. As a company, we have four
legs that we stand on. We’re an EMR, which stands
for Electronic Medical Record, data, which stands
for data, health information exchange in an API, Automated
Programming Interface, program company. We are built to be platform
and device agnostic. We’re built for Chrome. We believe in open APIs,
internet of things support. And we actually have Google
Glass support for image upload. Some of the milestones. We went live at my practice
in Cedar Park Pediatrics in February 2015, which
was quite a day for us. I remember with the
agile architecture and thin clienting,
we went live. And there were some
residual bugs left, but my co-founder literally
was changing things on the fly and we were moving and
moving and grooving. Since going live at
my practice, before we had kiddoEMR where we were
using another EMR client. It cost us roughly the national
average to process one patient visit, $58. We now pay $17.60. My doctors work four days a
week seeing about 25 patients where most practices work six
days a week seeing 40 a day. And our patients have
benefited for it. Between that time
and around May, we realized we had something. And we went and filed the
entity and started our business structure. And then I think
you guys may have seen me last October
[INAUDIBLE] ranting lunatic. By May we had founded
our company in London. And now here I’m
here to announce in our minimum viable product
for our patient portal and release candidate for
our verification trials. And we’re ready to scale. So it’s an exciting
time for us and you. My slides are a little
bit out of order. Does anybody have
any questions so far? So how does this work? How does the kiddoEMR
developer core, the Med Dev zone, and kiddoEMR and Unificare
and Dr. Joe and [INAUDIBLE], how does this all go together? Well, this first step is we
want innovation proposals and discovery, demonstrations. Some of them are going to come
out of the code kitchen today, I hope. Value add propositions
to our platform. Then we’re going to collaborate
together with the kiddoEMR developer core. And there’ll a link at the end
to join that private community. And what our physicians will do
will provide ethical oversight to the technology you’re trying
to implement in health care. Remember, in health
care specially, just because it can be done,
especially, with children doesn’t mean it should be done. We don’t want SnapChat for STDs. I’m sorry, kids. Sexually transmitted infections. Herpes, you know. That type of thing. I’m treading on
shallow ground here. Are any of your parents here? OK, good. Don’t tell them
you saw me today. But ethical oversight’s
very important. And that kind of
has to do with why I’ve moved over from Chief
Executive Officer to Chief Medical Officer. As a matter of
fact, the President of the American Academy
of Pediatrics on LinkedIn, he’s like, congratulations
on your promotion. Which I didn’t really
see it that way. I was like, but I
want to be in charge. But after seeing what
[? Raj, ?] our CEO in London has to deal with,
I’m like, thank God I don’t have to do that. And step three is publish. Publish. So what is the technology? This was actually taken
through glass in my office. I have permission to share
this, so don’t call the lawyers. So our technology. We talked about this. EMR data API and HIE. Let’s start with EMR,
Electronic Medical Record. Pretty simple. We basically are a data entry
portal with a HIPAA compliant database for the entrance and
maintenance of your health care information. Whose doctor uses an
EMR here, anybody? You need to find other doctors. Who’s writing on paper? They walk in, their
doctor writes on paper. Paper. It’s not bad. It’s pretty much the way
it is, unfortunately. It’s sad. Does anyone want to know
why their doctor’s probably still writing on paper? Anyone have any ideas? Yes. AUDIENCE: [INAUDIBLE]. JOSEPH COHEN: Funds. Smart kid. Here, take my card. Here, give him a pin. Give him a pin. I should probably take
them out of the bag first before I throw them. It’s aerodynamics. Physics was never
my strong suit. Yes, another one. AUDIENCE: [INAUDIBLE] JOSEPH COHEN: Excellent. Back up and contingency. Yes. AUDIENCE: [INAUDIBLE] JOSEPH COHEN: OK, so dictation
and dictation interfacing. Yes. AUDIENCE: [INAUDIBLE] JOSEPH COHEN: These
guys are rock stars. I don’t know any of
them, by the way. I didn’t ask them to come here. One more. On the right there. Both of you will have a chance. Don’t fight. Pick a first. AUDIENCE: [INAUDIBLE] JOSEPH COHEN: On paper? OK, what happens if
you lose it though? No indexing, right? Dewey Decimal system? Do you guys know what
Dewey Decimal system is? Well, I’m old, I know. And you would go to
the library and you would go through the Dewey
Decimal system, you walk in and the book’s not there. Then what do you do? Do you go to the search bar? The search bar at my
library served coffee. Anyone else? Yes. AUDIENCE: [INAUDIBLE] JOSEPH COHEN: Yes. All right, everybody
that answers something, you’ve all got jobs in our
new marketing department. So we’ve heard ease of use. We’ve heard archaic technology. We’re hearing my phone go viral. Thank you, everybody. We’ve heard expense, et cetera. Every one of you are correct. Absolutely correct. And as a doctor
making this company, I have done my very best to be
a domain expert establishing solutions that are credible. And this is what
doctors will tell you when it comes to why they don’t
like or are not using an EMR. Remember that the doctor usually
doesn’t have the decision. Usually they work for
somebody who says, look, we have this great new system. Let’s go for training for
a week on your Saturdays to learn how to use it. You may not know this,
but as physicians we’re horribly obsessive compulsive. As a matter of fact,
the obsessive compulsive personality type is 130%
more common in physicians than it is in the
general population. Why is that? Remember that if your guy that
you do your tires doesn’t fill a tire up, forgets to do it, you
walk back, you say tire’s flat. But if I forget to do something,
it could be life or death. And as a result, with
our OC personalities, we always do things by
the same procedural. Like for instance,
in the last 15 years, I’ve never approached a
patient on the left hand side. I was trained to do it on
the right, that’s what I do. I start here and I work my way. But in EMR, especially
those built by developers, are asking doctors to reproach
on the left hand side. They’re hijacking our process. And that is no longer
acceptable, in my opinion. They believe that
it can’t be better. They don’t have time to
learn how to develop. They don’t have time, like
myself, to show their passion and try to disrupt the industry. So we’re going to
do it for them. And we’re going to
show them that they can do it the way that
they’ve been taught to do it. If you hijack a doctor’s
process of care, they’re going to make mistakes. And doctors are really sick of
having to bend over backwards to accommodate their EMR. I’ve spoken to a lot of
my colleagues and even new doctor friends. The most common thing they
hear at a demonstration is when they make
a feature request, I want to be able to do A,
B, and C. The number one thing they hear from the
developers is, you’re not going to want to do that. This is how it works
with our system. And that, I’m sorry,
is unacceptable. In the era of Facebook,
in the era of Google, when we can track our
ex-boyfriend or girlfriend like a stalker and then we say to
the doctor, oh, but you can’t. You got to follow them by foot. The analogy’s thin, but very
important to understand. But we talked about, well,
the technology’s not the EMR. So what are we doing here? Well, it’s the data
that’s the real win here. By defacing data, it
becomes anonymized and becomes granular. It’s granular and it
becomes extremely pliable. And we can make conclusions
with that data that have never been able to be made before. Yes, question? AUDIENCE: [INAUDIBLE] JOSEPH COHEN: The
amount of money it costs to go through the system. Thank you. Thank you very much. I’m broke, don’t sue me. No, they got caps on it. It’s all right, I got my
stitch kit in the back. [LAUGHTER] I have tetanus shots too. All right, I’ll have
someone pass it back, Joe. Did you guys get it? Pass it back. Anyways, this is a really
overly convoluted slide that I’m going to skip. This one, though, however,
shows a little bit more of what architecture is like. Basically if you see this–
does that show up there? OK, so because
we’re in the cloud, because nothing exists in
the physician’s office, he can take his practice
based access or even remote access from home. I can see patients
on my way home, actually, if they’re
on the way, of course. Then we have a remote
hosting service where we use drive level
encryption and a private HIPAA cloud. And then around that, we
build this analytic cloud, which is the anonymized data. But you can also see here
that other people who have been credentialed
can access information that’s relevant to them with
granular access control lists. For instance, the school
nurse that is near my office. It is possible for her to
have an account into our chart so that her students, she can
have all of their shot records, any notes excusing them
from school, any medication authorizations. Who’s a parent? What do you have to do,
Stacy, when your school wants your shot record for your kid? AUDIENCE: [INAUDIBLE] JOSEPH COHEN: OK, Stacy’s
not a good– Stacy? No? OK, anybody else? Yes, what do you have to do? AUDIENCE: [INAUDIBLE] JOSEPH COHEN: Go to the doctor,
get it signed, go to school, drop it off. I didn’t make it up. Don’t look at me like that. But OK, you go to the doctor. Do they give it right to you? No. You have to then go to
the doctor, ask for it, go home, wait for a call,
go back to the doctor. AUDIENCE: [INAUDIBLE] JOSEPH COHEN: Which they’re
not allowed to mail it. Why? Whiskey tango foxtrot. Why? Why? OK We’re going to change that. And we’re changing that in
Cedar Park in our microcosm. Our school nurses will
log into the system. We will have our patient’s
consent for that ahead of time. And they will have the
health information exchange, we’ll give them the
information they do. And you know what? They won’t bother my staff. It takes a lot of time for
us to process that as well. Open APIs. Open API project. Everyone know what an API is? Could someone explain it to me? Never mind. This is a quiet crowd here. We’re going to harness the power
of a full open source platform. This whole room snafu got me
all turned upside down today. We’re going to integrate
third party services. Let me give you an example. We have a partnership
with a company of Paris called Slow Control. They are about to
announce their Baby Gigl. I didn’t name it. And the French spelled
it G-I-G-L. So Gigl. What it is is a sleeve. It’s Bluetooth enabled
with an inclinometer and volumetric sensing. So who has a baby? You need to keep
track of feeding. Pour it in the bottle,
you measure it, you put it, write it down,
give the baby their feed, write down how much they
ate, how long it took. What this sleeve
does is it actually knows how much you
pour in the bottle. It senses the parent the
best angle to hold the bottle and it evolves as they feed. And also records volume,
lumps, air feeds, rate of feed, and presents it in
a graphical format. So when a patient brings their
newborn to me, I sit there and I say, OK, how’s
the feeding going? Good. All right, we’re going to go
and take a look at him now. Instead with our Baby
Gigl, all the feeding data that’s been compiled
by the hardware will show on the date of
service and give the doctor not only volume, but
pattern, rates of feeding, caloric intake, and overall
calories per kilogram per day with no effort at all. And because we can stream
that data through our health information exchange
using our APIs, we can present that data
to the doctor in real time. And the parents are
more engaged as well because they know
they’re getting the data. So we’re either doing with an
over-exhausted newborn parent who’s like, it’s going good. Or we’re dealing with a
platform, a modified self platform, that gives
us the data securely, privately, and efficiently. And I think that
works out better. There was a question over here. Please. AUDIENCE: [INAUDIBLE] JOSEPH COHEN: Yes. AUDIENCE: [INAUDIBLE] JOSEPH COHEN: That’s
a good question. And My co-founder is here and is
going to talk a little bit more about that in a few minutes. And I want to stick with
that question in a second. But what the answer
is, it does have to do with end user
authorizations as well. So your level of security
is my login to the platform, timing out, auto timing out so
that nobody else could walk up behind you and see it, logging
in on your device and whatnot. Also the fact is we don’t
sync to our workstations, we expose to our workstations. But we’ll have more on that. I didn’t leave any
of that out, Ron? I left some out? I got it. My talk last year was titled,
I’m A Doctor, Not A Developer. Yeah. Let’s see some of
these things in action. Well, yeah, tears, I know. OK, let’s talk about
camera support. These are two diagnostic images
I took with Google Glass. These are images that are
actually in patient charts. And patients who have
doctors that write or use most conventional
EMRs, this finding would be documented as, this
one on the here, this one is two flesh colored keratinized
papules with central puncta ichthyosis of one to two
millimeters on the extensor surface of the right elbow. How long does that
take this type? It takes about 45 seconds
to a minute for me. Students maybe two to
three, five minutes. Instead I take the photo,
it shows up in the charts, and we move on. So I’ll annotate here instead
voeller surface right elbow. And the other one. Large macules of
alopecia consisting of vellus hairs and exclamation
point hairs of 25 to 30 millimeters on [INAUDIBLE]
without any signs of [INAUDIBLE]. Or we can, say it with me. AUDIENCE: Take a picture. JOSEPH COHEN: As
a matter of fact, we’re in for a treat today. I’m actually going to show you
our demonstration server, how that works. So these are mock patients. Let me hope that that’s not–
is that the minimized one? Sweet. So in a patient’s visit. We’re going to actually use
just the device hardware, because this module requires
us to be on a secure network. We have an image assessment. Whoops. Oh, is it going to do that? There we go. So our image [? grabber. ?] You
can actually capture the photo. It’s going to be a little
slow because the network here is so fast. So here we are. I’m capturing now. This camera can be anything. Could be your camera
for your cell phone. Could be Google Glass. It could be something
like the HTC RE camera. And we capture it. I look great, don’t I? 43 tomorrow. We capture it, it takes
forever because we’re not on our network. It goes into the
date of service here. And I can annotate. Going to be 43. Save it. And now the
information’s showing up on the date of service. And I’m going to show
you a little bit more on this, a few examples on how
that actually means something. So back to that. Let’s take a look at
this boy right here. With the direct
camera support, we’re preserving the visual diagnosis. Does this mean anything to
a parent of a sick child? Does it? It means a lot. We have a lot of teens here. Your acne, the doctor’s writing,
not that any of you have it. Thank you, good night. [LAUGHTER] STDs and [INAUDIBLE] acne. That’s great. We’ll be having another
talk at two o’clock where we’re talking
about tech in medicine. I’m sure you guys are going
to love to– kids in tech. Anyways, if the doctor wanted
to know how you improved, wouldn’t it be better if you
had a picture of your skin to compare with? This is a very unfortunate
boy who had something with a really long name. And the top picture is when
he came to see me first. So then I did some of my magic. And you look at
the bottom picture and that was about
a day or two later. Is there any doubt in your mind
that this patient is improving? You all have
honorary doctorates. You just did it. You just did it. Is there any doubt
in the lawyer’s mind of the parent later on that
says he didn’t make them better? Is there any doubt? Because I’ll tell you, you
know what the chart would say? It would say two flesh
colored keratinized papules with central puncta
ichthyosis of one to two millimeters
on the extensor surface of the right
elbow, dash improved. That’s what they
train us to do, OK? That’s our standard
of care, people. We’re going to change
the standard of care and enforce it. It lowers your liability. It helps you gauge progress. It’s meaningful documentation. It takes no time, it’s secure,
it’s safe, it’s sufficient. It improves the health
and lives of my patients. My patients need that. Here’s another example of how
we use our APIs through a HIPAA compliant Google
Calendar support. Yes, with my locked phone I can
have my HIPAA compliant patient calendar on my Google
Calendar with full compliance. Try to do that with one of
the large corporate products. Not going to happen. Shows the power
of our open APIs. Our Google Glass support. This is a wire frame that’s
really small and blurry. But it shows how we can use
glassware, the wonderful mirror API SDK, whatever it’s called. One of those things. Both. It’s SDK. And we can use the hardware
to capture meaningful data for our patients. Any questions? Health information exchange. This is a generic– yes? AUDIENCE: [INAUDIBLE] JOSEPH COHEN: Oh, it needs work. That’s why I’m
talking to you people. It needs work. I’m trying to find it. There it is. I used to wear Glass a lot. It’s kind of getting
past my threshold. I can’t wear something
three years old. But it’s really slow,
it gets really hot. It’s at work where I need it. But we need to see
improvements in the hardware. We need to see other
form factors that don’t make you look like
someone on “Star Trek.” Actually they look
better on “Star Trek.” But oh, you just
filmed that, Google. That’s from Google. I’m going to hell. There was a question back there. Yes? AUDIENCE: [INAUDIBLE] JOSEPH COHEN: OK. That’s a good question and
I may have not been clear. It’s the doctor’s responsibility
of taking the visual diagnosis and actually writing
what he thinks it is. In the medical data format,
I’d take the picture and in the assessment
section of the plan. So we have a child
with that rash. That first one is
actually just some warts. So I would actually
assess it as warts and write it in
my treatment plan. Unfortunately, we cannot make
technology legally in this country that acts like a doctor. We can only make technology
that assists a doctor. Otherwise you’d put
me out of a job. Health Information Exchange. The universal adapter for the
efficient streamlined transfer of HIPAA compliant data,
no matter what it is. It’s all ones and zeros. Thank you. Am I sweating more now? Did I pee my pants? Oh, sorry. And it can be used for
any amount of data. We’ve laid the
infrastructure down. We’ve established the personal
data transfer protocols. And this is where all of
it gets really exciting. This is our patient
portal prototype. And what it allows you to do,
what it allows our patient to do, more importantly, is to
completely manage their child’s record on their phone. This is an exposure framework. So basically they log in and
the data opens, very similar to, say it with me, Facebook. If you’re not logged
in, what’s there? Nothing except for the stuff
they put on there to track you. We’re not doing that. This allows a parent to check
their upcoming appointments, reschedule set appointments. Allows you to read
messages from your doctor. It allows you to pay for any of
your services and co-payments. Also allows you to
refill your prescription. What did you used
to do before this? What did you do? You went in, you
stood in line for them to forget you and go
back four days later and go back and stand in line. No more. The other things
that patients can do is check out their
lab test results. This is a mock up, so
it’s not very clear. Medication, allergies,
medication list. So let’s go back to that. Let’s go back to that example. Parents are out of town,
you didn’t listen to them, and your friend ran over
your foot with their car. That’s actually a true story. And you have an allergy to a
medicine they need to give you, but you have no idea what it is. You show them exactly
what it is here. It saves time, saves
lives, saves money. Yes, in the back. AUDIENCE: [INAUDIBLE] JOSEPH COHEN:
Eventually it will. It doesn’t. He asks if it has it access
to insurance information. It’s not in the best interest of
large health care conglomerates to expose their data. So there’s tons of
different systems. They’re not all meeting
the same convention. But eventually they’ll
see it our way. And if they don’t,
then the patient can help advocate for
their own data transfer. A patient can do whatever
they want with their data. But that’s a great
question and we are going to just have to
disrupt the industry until they see it our way. Yes? AUDIENCE: [INAUDIBLE] JOSEPH COHEN: That’s
a good question. And that has to do with the way
that the database is encrypted and also in our
hardware infrastructure. And we’re going to get a little
bit more into that later. Most of the data breaches,
as a matter of fact, all the big high
profile data breaches that you’ve read in
the news, they all had their data stored
with no encryption. In Health was one. Someone just walked up to the
server with a thumb drive, stuck it in, took their data. So also reports, whatnot. Let’s go back to
this presentation. It’s exciting, isn’t it? And it’s going to save us time. So imagine a patient
has a picture from the rash from last night. They can actually take the
photo stored in the app and the doctor pulls
it out of the cloud and puts it in their chart. So the patient came in. This is what it looked
like last night. This is what it looks
like today in my office. And then tomorrow, this
is what it looks like. Yes, in the back. AUDIENCE: [INAUDIBLE] JOSEPH COHEN: It’s
completely personal. Nobody random will
able to see it. You need to
authenticate yourself. You also need to identify that
the doctor you’re seeing with and authenticate them. And they would see it. A lot of people
worrying about security. That’s good. So why Android? Well, Android, I see most of
the phones I see in my office are Android headsets. Cracked Android headsets,
because kids just are like, oh, great. It’s like a splinter
waiting to happen. It’s really good for business. But it’s the most affordable
and generalizable platform, say it with me, at scale. Thank you. You said it a little after me. Let’s work on that. And children are our number
one renewable resource. [LAUGHTER] What? They are. They are. What? OK, cellphones. Did I say something wrong? Special considerations
to health care. Affordability, accuracy. We don’t want to
[? portray ?] our information with less accuracy. Privacy we’ve talked about. And also one thing that
we don’t always talk about is accessibility. You may not see a lot of
children in wheelchairs at the mall, but you
see a lot of children with wheelchairs in my office. What does that tell you? That your audience is more
likely to be a specialty audience. And you better bake in some
accessibility for them. Because they’re just as entitled
to this revolution as we are. So keep calm and follow
the prime directive. HIPAA, Healthcare Insurance
Portability and Accountability Act, which really has nothing
to do with portability at all, is all about keeping data
private, keeping it protected. It’s the prime directive
of health care information. And we’re going to keep going. So oops. I’m going to do so I
co-founder, Ron Criswell. He’s the tech guy. And he’s going to take a few
minutes to talk to you guys about how we’ve established our
HIPAA compliant architecture. Please welcome Ron. Ron Criswell. RON CRISWELL: Try to answer
some of the questions that people brought up about
how we have implemented a lot of these
tasks and how we’re going about making the software
and the data accessible and still maintaining security
and protection of that data. We have kind of two
goals that seem to be opposite ends of the spectrum. We want to make a fairly
open, easy to use, fairly simple
platform and API that encourages development and
encourages people to use it. At the same time,
we need to make sure that we adhere to all of
the HIPAA requirements, that we maintain
security, that we maintain privacy in that data. The way we’ve gone about
doing that is we use fairly standard technology, browser
based technology and supported HTTP protocols in a method that
is compatible with application development on either a portable
platform with an Android type application or with HTML
inside a browser like Chrome or Firefox or
something like that. Essentially what we do is
you have to authenticate. Your authentication
mechanism from the API is OAuth 2, which is a very
familiar, very standardized mechanism that Google uses in
a lot of their applications. What OAuth 2 does
is it allows us to validate who the
device is or who the software is trying to talk
to us as a primary connection security. And secondly, it
allows us to verify that you have access
to the specific account that you want to have access to. Somebody was asking
earlier about how you make sure
that you’re only talking to this one person. That’s the way we do it, is we
provide the OAuth 2 security algorithms that allow an
application to connect to us. And if you provide the
correct credentials, then we’ll allow
you to access it. It’s the exact same
kind of technologies that the banks use for signing
on to your online banking accounts. Very tried and trued
application, very secure use of the whole algorithm
and schema process to be able to do that
kind of connection. So that’s one part of
our connection process. The second part is we’re
using an SSL connection. End to end communication. We never provide a mechanism
for storing the data on the end point. You don’t stored it
in the phone app, you don’t store it
on your computer, you don’t store
it in the browser. It’s always a pull the
data, display it, and then throw it away. That’s necessary to make sure
that you don’t have problems with storing unencrypted
data anywhere in the set up. Yes? AUDIENCE: [INAUDIBLE] RON CRISWELL: We’re not doing a
two factor authentication right now. We could do it. And we’re kind of
looking at that as an alternative for
future mechanisms. Like I say, right now
what we’re doing is we’re checking to make sure
that the application that’s trying to talk to us
is our application or it’s an approved application. So the application has
a set of credentials to authenticate itself,
And then there’s a secondary level for
the authentication of the individual records
that they’re trying to access. That’s as far as we’ve
gone with it right now. AUDIENCE: [INAUDIBLE] RON CRISWELL: Not at this
point, but that’s a good point. And that’s something
that we may need to go through and
do something more along the process of having
an identifiable figure that a lot of the
banking applications use, and some of those
types of things. These are the types
of areas that we’re needing participation,
we’re needing input. We don’t pretend to
have all the answers. I’m kind of giving you
where we’re at right now. But we definitely have
room for improvement. We need input from
the communities. We need to know what types
of things are recommended. That’s what we’re looking for. That’s what we’re hoping to
get from our partnerships with developers are, what
are the areas that we need to improve on? As far as the data, we keep
all of the data encrypted on fully encrypted databases so
that there’s no chance of being able to access it brute force. But there is, when
you’re transmitting data, you do have that vulnerability. You need to make
sure that you’ve got the correct connections. And somebody else
had a question? So anyway, we’re
starting to more people in, so I’m sure
that we’re getting close to the next
presentation [? side. ?] But really what
we’re doing– oh, the other piece for
external access, the API access, what
we’re working with is standard RESTful APIs. The data connections
are designed to be able to provide access. What we’re looking for in
help in the development arena is for people to come up
with new and interesting ways of processing that data
and making that data available to other applications
and in ways that help improve the quality of the health
care, help improve the outcomes and so forth. We’re providing the
tools into the data. What we’re looking
for is help coming up with better, new methods
of presenting the data, massaging the data, and
making the data useful, turning it into information
for the end users. JOSEPH COHEN: Thanks, Ron. Thank you. That clear it up a little bit? A couple of things right now. We do have stopgap
measures for two factor. Any of our cloud
based access, you need to authenticate through
VPN before you access it. And then our actual native
site of practice, you actually would have to have the
credentials for the network as well. So we’re very well aware
of– excuse me one second. We’re very well aware of
the need for two factor authentication. I think that it’s
very important. We have other ways of
solving it using two factor and putting our login
wall behind maybe Google App for Business,
Google Apps for work where you’re actually being
enforced two factor that way. But one of the things
that we’ve been looking at with the mobile
technology, how does the patient link to the doctor? I feel that we use kind
of like a Bitcoin model where you scan a
code or you enter a code that’s unique
to that physician that you can only get there at
that office that connects you through the application to
your chart and your provider. But these are
excellent questions and exactly why we’re here. Because we need your help. We need your help
to take our platform and present it to our
patients and hopefully your patients that you know
and people they know as well. And this is the process. Basically you’re going to
come to us with a discovery or proposal. You’ll present your ideas to the
develop core or to us directly. We’ll then go and vet it
with medical and ethical advisory board that we already
have in London and here in Austin, here in Texas. We’ll actually analyze its
practicality, the ethics, and if it is indeed
apps with impact. Then we’ll go through prototype
and testing, and analysis for HIPAA compliance
and usability. And finally, release of an MVP
with reiteration and update process. It’s time, guys. We can innovate, disrupt,
save money, save lives, make apps with impact. Actually this right here
is the London store, the Google store in London. If you ever there it’s on Warren
Street in Tottenham Court Road. It’s a lot of fun. They have a little
spray painter. It’s a digital
spray paint bottle that you can actually pick
your color and spray paint. And course in true Google
form, when you shake it, it makes the sound like little
balls running around in it. It’s pretty neat. I go there every time I
go, and more than once. So if you’re interested
in working with us, this QR code will take you
to the devoted to the dev corp page. And you can also
search it on Google+. Please join. It’s a private community
because we just want devs. Devs and kids. And people that are interested. We just don’t need trolls
and people that are not relevant to the project. This is not a bridge in Denmark. You can also hit
me up at my email, on Twitter @ikeysee or at
kiddoemr and then +joeycohenMD or +iamdrjoekiddoemr, because
that makes a lot of sense. And I’d like to see if anyone
else has any questions. Did we answer them all already? OK, I have one more Seattle pin. Does anyone have a question now? [LAUGHTER] Yes, he has a question. AUDIENCE: [INAUDIBLE] JOSEPH COHEN: I’m going to
reiterate your question. It’s a good one. What he asked was, there’s a
lot of EMR systems out there. Is there a common framework
for transferring information? Because the default
right now is fax. And yes, I do, I have a
fax machine in my office. My name is Dr. Joe and
I’m stuck on fax machines. It’s disgusting, isn’t it? Absolutely. And there are. There’s the CDA, Clinical
Data Architecture, protocol, HL7, Health Level 7. These are open
source standards that are prepared to [? handle ?]
the transfer of data. kiddoEMR, we just finished
our C-CDA architecture import export. We can actually take
information from anybody’s EMR and incorporate it
into our platform. Unfortunately, most
EMRs don’t want you to do that, which in my mind
is corporate greed interfering with the health and
well being of children in this country and others. And so we’re using
open protocols to help facilitate that
data transfer as well. Here. I know. We can edit that out, right? Yes? AUDIENCE: [INAUDIBLE] JOSEPH COHEN: Good question. FDA approval. What is the threshold
for something to become FDA approved? Yes. I talked a little bit
about this last year and I’m glad you
brought that up, because I’m realizing I had a
few holes in my presentation. I’ll get to you. You can put your hand down. When you go see your doctor
and he’s not using an EMR, he’s using a what? Pen and paper. Is a pen and paper FDA approved? OK, let’s say he’s using an EMR. Is his MacBook FDA approved? Is his EMR FDA approved? No. But if he builds a
machine that tested blood sugar, an actual medical
intervention, and that affected the treatment of
that patient, he would have to have
that FDA approved. There was an app that
was a diabetic app that was built that was proved
to lower blood sugars better than Metformin, a drug. Well, that had to
be FDA approved. And it was actually
nothing in the app that was actually a medicine. It was just dietary
modification app. And that needed to go
through FDA approval because it was
actually implementing a care plan for the patient. Yes, you want to follow up? AUDIENCE: [INAUDIBLE] JOSEPH COHEN: If
you were building, you would need to go through
the reliability of the device to actually tell you
what is actual the truth. And that would require
the FDA approval. But let’s say you were building
a Chromecast for an app where you can project. And that probably
wouldn’t need to be. I’m sure that there’s
gray areas in this. And this is a great question. I’d like to talk to you
more afterwards about that. And really what he’s asking
is, where’s the line? In other words, where does
the line between FDA approval and not needing FDA approval,
where’s that line drawn? And right now it’s drawn
through medical intervention and evaluation and management
tools and treatments and therapies. Does it mean that we can’t do
a handshake, a data handshake? I’m not sure. That’s a good question. Who else? Yes. AUDIENCE: [INAUDIBLE] JOSEPH COHEN: I’m sorry,
I didn’t hear you. Can you stand up
and say it louder? AUDIENCE: [INAUDIBLE] JOSEPH COHEN: That’s
tough over there. I didn’t think about that. Go on welfare? A lot of my family,
when they see on LinkedIn being CMO of this
and doing that, they’re all like, congratulations. We knew you’d be successful. I’ve been running a
practice 15 years. That’s my success. And if we were to go massive
electromagnetic pulse tomorrow and every computer
was died and fried, I’d just go back and
see my patients again. But as far as you
guys are concerned, what we’re trying to do
is build an ecosystem. We’re trying to build a standard
for the care of patients. And the cloud and
the architecture is there to satisfy
the patient’s needs long after I’m in the ground. There’s a famous quote. If you meet your goals
or you set your goals we met by the end
of your lifetime, you aren’t thinking big enough. And really what I’m trying
to do here is just the start. So that’s good. And hopefully with great minds
like you all contributing, we won’t have to
worry about failure. Yes, in back. AUDIENCE: [INAUDIBLE] JOSEPH COHEN: It doesn’t. But it does. OK, that’s a bit weird. It does. That’s our plan. The data exists. We have a very robust tenured
historical granular data. But that is exactly
what we’re trying to do. I could show you one
thing, if you want. I did a heat map. This I did about
a year ago myself. And I did a heat map. I use a weather
mapping algorithm. And I took my data from–
can you guys see that? I took my data from two
winters in Austin, Texas. I took patient location
and time and severity of allergies in January
and February in Austin. That means you’re
allergic to cedar. And I took severity,
location, and time. And this is what came out
for the months of December and January of 2013 and 2014. If you see here, the red is
the most severe cedar fever allergies, and green
and yellow are less. What this actually showed us
is this green belt right here was pollinating all
of Lago Vista, Texas. And the mayor of
Lago Vista actually was very interested
in this analytic and we actually got
some press off of it. But that’s what
we’re talking about. Robust data manipulation. Presentation in ways
that actually matter. Now, I’m a human
who has a day job, and this took me
about two weeks. But with the proper
architecture, the proper programming,
the proper brilliance, we’re working to have this
done autonomously, credibly, and in real time. So I really like your thinking. You definitely need
one of my cards. Yes? Yes? AUDIENCE: [INAUDIBLE] JOSEPH COHEN: She’s
asking, is this going the requirement for
all doctors in health care professions to use it? No. In England, the
National Health Service has tried to make something
a requirement for doctors in England in 2012. They spent $15
billion on a solution and had to mothball it
because doctors don’t like being told what to do. Go figure. We tell you what to do all time. And our solution is bottoms up. Already the cost
savings is making it pretty undeniable argument. But why do people use Facebook? Because they want to
have that convenience. They want to know where their
high school friends are. They want to be able
to share pictures from when they were kids. They want to use it. Doctors want to use this. It’s just we need
to get it to them. And it’s going to
take some time. Who else? Am I missing anybody? All right, well thank
you very much for coming. I want to thank everybody
for paying attention. [APPLAUSE] I’ll be here all week. We have at two o’clock a
kids and technology fireside chat in room six. And also the big barbecue’s
coming up tonight. I don’t know if you
guys have had Ray’s. It’s amazing stuff. So thanks for coming. [MUSIC PLAYING]