PHI From Glass

A handful of surgeons have live-streamed surgeries to the Internet from Glass using Google Hangouts. Although in each case, the patient consented, using Google Hangouts isn't a viable production grade video streaming solution. Google Hangouts aren't encrypted or secure, and the videos will live on Google's servers forever.

Sending any PHI through Google's servers, including every application that utilizes the Mirror API - is a HIPAA violation by definition. Using the Mirror API, Glass will never be a viable solution in healthcare.

However, there's another way. At Pristine, we're writing native Android applications that reside on Glass itself, and that communicate with our own server platform that'll be installed within the hospital's firewall. We encrypt data in transit and at rest, and we generate a complete audit record along every step of the way. And most importantly, no PHI will leave the hospital's premises. Hospital management will control and manage data that they capture on their own terms, not Google's terms.

Every viable production grade healthcare application on Glass will have to be written as a native Android application. As such, the vast majority of innovation over the past few months on Glass for healthcare has been just for show. The Mirror API is fine for rapid iteration and testing screens, but will never be usable in production healthcare environments.

Ephemeral Communications

This post was originally featured on HIStalk. 

SnapChat is ridiculous. It’s raising $80M at an $800M valuation. How are self-deleting pictures and videos worth nearly a billion dollars?

SnapChat discovered a nascent human behavior: people share more stuff when they know it can’t come back to haunt them. I thought SnapChat was only for sexting until I started using it. One hundred percent of the SnapChats I receive are of my friends being idiots throughout the day. They’re funny and silly, but usually a bit embarrassing. I never received half a dozen silly selfies via traditional texting. But with SnapChat, I receive silly selfies all day long because my friends know I’ll get a kick out of it for 10 seconds and then forget.

Healthcare could learn a thing or two from SnapChat. The healthcare system discourages providers from communicating (in order to care for patients) so that there’s someone to blame if things go wrong. The hallway consult is dead. Providers should communicate whenever and however they need to in order to get the job done.

Doctors shouldn’t be allowed to abjure data from the EHR. The EHR should have a complete audit record. But providers should be allowed to communicate without fear of legal repercussions, when and where appropriate.

More broadly speaking, this is a question of communication channels and intended audiences. Some communications were intended to be strictly provider-to-provider and were never intended to be provider-to-patient, provider-to-lawyer, or provider-to-insurance. In an ideal world, the EHR would be strictly provider-to-provider. That would allow providers to communicate on their own terms.

Each of these non-provider entities expects that everything in the EHR should be written for their needs: patients want patient-friendly data and notes, lawyers want to see as much documentation as possible (no matter how irrelevant or the cost it places on other providers that have to wade through it), and insurance companies want as much documentation as possible so they can find a reason to reject claims.

Providers can’t create multiple versions of the EHR in which each version written for the non-provider parties that want to read and access that data. That would simply be too time consuming and laborious. Patient portals have begun to tackle this issue, with links from diagnoses and lab results to websites that provide a medical explanation in a patient-centric fashion. But patient portals still have a long way to go.

Other initiatives, such as OpenNotes, have been received positively by providers and patients, though I can’t help but wonder what information providers have started leaving out of notes knowing that patients will read their comments directly.

A handful of companies, such as Avado, are trying to tackle this issue head on for patients. Some specialties are better suited to this than others. For example, Avado has been particularly successful with expectant mothers  who are excited to learn and engage with their providers. On the other hand, patient participation in ophthalmology is poor since many patients are older or can’t see well.

Perhaps with enough data, someone can create a Google Translate for the various parties that want to access EHR data. Please email me if you have any insight on this.

The Killer Apps for Glass

This post was originally featured on the Pristine blog. 

I recently wrote that there aren't any killer consumer apps for Glass that will drive adoption. Yes, there're Glass apps for hobbies, but those won't drive mainstream adoption by 25% or 50% of the population.

But Glass will change how a lot of people do their jobs. To understand that, let's consider the marginal value of Google Glass:

1) hands free

2) heads up display

3) friction free

4) first person camera

Glass is particularly suited to help individuals who have the following job description: folks running around all day using their hands who need to access and share information. All of the following job functions match that job description:

medical professionals

security / surveillance / police

retail

warehouse workers

restaurant staff

on site technicians / engineers

For people who have that job description, Glass solves problems that are present for 8+ hours / employee / day. That means that in these scenarios, Glass has the opportunity to create enormous amounts of value.

The killer apps for Glass will be for the enterprise.

What's the Killer App for Glass?

With the exception of the iPad, the adoption of every major consumer technology platform was driven by a handful of distinct, killer applications:

Desktop: Excel, later Word, and then the web

Cellphones: voice + text

MP3 players: music

There have always been niche applications, such as Photoshop, or AutoCAD, for specific verticals. But those verticals never drove adoption of the technology platform by 25% or 50% of the population. There have always been just a few killer apps that drove adoption.

One might recall Apple's series of commercials, "there's an app for that", over the past few years and say that the killer-app theory is wrong. But that's not quite right: people bought phones for voice and text; Apple highlighted the apps as a way to differentiate themselves relative to Android.

But what about the iPad? How did Apple drive adoption without a single unique iPad specific application upon launch? The iPad handled all of the common computing tasks, but was cheaper and easier to use. Additionally, the iPad drove touch based computing. Large touch canvases are powerful and intuitive. They are inherently natural to use and manipulate, especially relative to a mouse and keyboard. No baby every figured out how to use a laptop on their own, but there're dozens of videos of babies using iPads with 0 training. Touch, simplicity, and better/cheaper/faster were the defining characteristics of the iPad. The iPad was successful because it made more computing more accessible to the masses across virtually every dimension.

So what about Glass? What's going to be the killer app that drives adoption by 25% or 50% of the population? There isn't, and won't be one. The problem with Glass is that it's not actually better than the status quo for the vast majority of the general public.  Glass competes with the smartphone, and it's materially worse at all of the major smartphone functions: voice, text, Facebook, Maps, email, and web browsing. Unlike the iPad, the user experience is quite unnatural. Given the screen size constraints, the absolutely wretched awfulness of the trackpad, and the timeline, the entire UX is simply unintuitive. I've personally watched at least 500 people put on Glass for the first time across a variety of social contexts. No more than 10% were able to understand Glass's UI without guidance.

Moreover, marketing Glass is an uphill battle. It's easy for Apple to make commercials of someone's hand swiping away on the iPad. But there's no way to virtually simulate Glass. You have to put Glass on your face and play with it to understand. And it's not immediately obvious how to use it. Sure, Google tries to provide virtual simulations, but these kinds of simulations aren't nearly as accessible as a hand swiping across a beautiful  touchscreen.

For Google's sake, I hope Glass succeeds as a consumer product, I just don't see how it will. Glass needs a killer app, and it's not there.

The Pristine Story: Building Infrastructure

This post was originally featured on the Pristine blog. 

The past few weeks have been all about building and investing in Pristine's future across virtually every dimension of the business: talent, work environment, sales infrastructure, marketing, PR, and capital. The pistons are firing:

We were accepted into the Capital Factory (CF) Incubator program, the most prestigious incubator in Austin. We're right in the heart of downtown, on the 16th floor of the Omni Hotel at 7th and Brazos, just one block from Dirty 6th. The CF team has been incredible to work with. I need to send a big shout out to Josh Baer, Gordon Daugherty, and the rest of the CF management team.

We've raised $350,000. The remaining $100,000 has been a bit slow, primarily because we've been looking for the right angels. We're about to finalize some very strategic angels to close out the round. I've also started planting seeds for our series A. As I've said before, this is a gold rush business, so we need to raise as much money as possible as soon as possible. With a couple of weeks of pilot data, I'm going to be making the rounds to raise a hefty series A. That means I'm going to spend virtually all of October and November begging for money.

I've been telling folks that raising money was easier than I thought it was going to be, and that hiring was more difficult than I thought it was going to be. After pouring through more than 3,000 LinkedIn, AngelList, and GitHub profiles and calling over 100 engineers, we've assembled the technical team that's going to make this happen. Our internal technical team is comprised of 4 engineers: Patrick Kolencherry (CTO), Mark Troutfetter (VP Engineering), Arik Yaacob (Sr. Technology Architect), and Rahul Behera (Mobile Developer). I'm especially proud to welcome Rahul aboard: he was smart enough to leave the University of Texas to work at a startup. That takes courage, and tells us that he's going to do whatever it takes to kick ass. If you throw yourself out there, you'll find a soft landing.

Additionally, we've hired 2 outsourced developers. In total, we have 6 engineers working more than full time. We're also very close to bringing aboard a Director of Sales, who'll be instrumental in closing our first customers in October and November while I'm on the road raising capital, and an Active Chairman. Although the core team is assembled, we will hire any rockstars that believe in our mission. There're always talented folks out there looking to work on incredible projects, and we'd love the opportunity to work with them if this is something they'd like to be a part of. If you know anyone that would want to be part of Pristine, please don't hesitate to hand out my email address.

We also updated our website. Check it out. Thanks to Bryant Peng for putting in the leg work to make this happen.

I continue to live life as a road warrior. I went to San Fransisco and New York for investor and partner meetings, Las Vegas for SXSW to network, and I just got back from a visit in Atlanta where I gave a presentation to an executive team of a major healthcare IT company. Surprisingly, I don't have too many more travel plans until the end of September (Health 2.0 + GlazedCon in SF), though I suspect I'll end up traveling over the next few weeks as new things come up.

Looking forward, the next month is all about execution, as we prepare for our first pilot site to start using Glass in the OR with live patients in September. The single greatest barrier remaining is lack of Glass. We have 3 units, and we have a few more coming, but we could use as many units as we can get our hands on. If you know anyone that would be willing to loan a Glass unit to a hospital to try to save lives, please let us know. It would mean the world to us. Unfortunately, Google hasn't been particularly accommodating.

Save the date: October 16th. We were officially invited to launch on stage in front of the world's premier VCs and media outlets at the DEMO conference in Santa Clara. I'll be on stage demoing our products live, and DEMO will live stream everything. I'll send out details as we get closer to the launch at DEMO.

And of course, I continue to blog religiously. The past few weeks have been the most prolific writing weeks of my life. Since I take care of most of my writing at 38,000 feet and because I've been flying so much, I've had the chance to do a lot of writing. Enjoy.

HIStalk

The Power of Connectivity

Battle of the App Stores: Athena vs Greenway

Computers are Eating Healthcare

Follow the (Clinical Trials) Money

There will Two Kinds of People in the World...

HIStalk Interviews

Paris Wallace, Founder, CEO, Ovuline

Joe Reinardy, Founder, CEO, CenterX

Josh Stein, Founder, CEO, AdhereTech

My Blog

Answering the Existential Startup Question

I'm a Professional Beggar, Storyteller, and Fortune Teller

Your Margin is my Opportunity

Is Glass a G+ or a Wave?

Engineers are King

Scorch the Earth

Blogging and Signalling

The Genius of the Glass Explorer Program

Selling a Dream

A Day in the Life of a Startup CEO

Understanding the ACO Pioneers Successes and Failures

Glass <3 FDA + HIPAA

Explaining the 17% Meaningful Use Dropout Rate 

PS, our Oculus Rift arrived. We'll have videos of that and of our offices at CF in the next issue of the Pristine Story.

Cheers!

Kyle, and the Pristine team