Tuesday, April 26, 2016

Getting on the Blockchain Bandwagon

Face it: health care IT infrastructure is a mess.  After spending tens of billions of dollars to "incent" providers to move to EHRs, they're using them but are not very happy with them.  In a world in which health IT systems should help improve patient care, they're seen more as a burden than as an asset.  We now have millions of electronic records that are still way too siloed, and all too often incomplete.  Worse yet, when those records aren't being hacked, they are being held captive by ransomware.

Enter blockchain.

To the extent most people think of blockchain at all, it is in relation to one of its most prominent users, Bitcoin.  Bitcoin, which has its passionate advocates and equally passionate skeptics, is not synonymous with blockchain.  Blockchain is the technology that allows Bitcoin to operate, but they are no more one and the same than Salesforce.com and Oracle are.   

In layman's terms -- and, trust me, when it comes to this I definitely am a layman -- blockchain is a set of distributed records, or databases, that are shared by multiple parties and which can only be updated by a majority of those parties.  There is no central authority, no central database.  It reminds me of the Internet's distributed networks, which help assure its robustness.

Equally important, in blockchain once a record is stored (or "transcribed"), it can't be tampered with.  For better or for worse, Bitcoin has demonstrated that blockchain does, in fact, assure anonymity, privacy and security.

Blockchain is starting to become more visible even outside of Bitcoin.  Businesses are being told they need a "blockchain czar.  Wall Street is starting to embrace it.  Britain looking into using it for manage the distribution of public money.  The EU is being urged to resist regulating it too soon.  DARPA wants to use it for a secure messaging system (which is how we got the Internet, as you may recall).
Some people think it is the greatest thing since sliced bread -- or, in modern terms, since the Internet.  IBM's Jerry Cuomo says: "Blockchain has the potential to become the technological foundation for all electronic transactions conducted over the Internet."  Similarly, venture capitalist Jalak Jobanputra gushes:  
Once every 20 years or so, new technology comes to market that has the potential to change how we communicate, how we do business, and pushes the boundaries of everything we thought was possible.
In the 1970s, it was the personal computer; in the mid-1990s, the commercial Internet. Now, we are in the early stages of another major technological development called the blockchain.
If they are even remotely right, blockchain is something that we better be paying attention to, and what industry needs its advantages more than health care?

Think of it this way.  Health care evolved in a world where the data, and the transactions, were initiated in the providers' office, made their way to the health insurers' offices, where they were processed, and sent back to the providers, which then updated their records and billed their patients.  Everyone gradually added computers to their part of the process, and started communicating electronically -- ultimately using the Internet -- but the underlying records and processes aren't much different than their 1970's/1980's/1990's predecessors.  Sure, the records are more often electronic, but everyone holds on to their data zealously and shares access to it only grudgingly.

It is thus no surprise that interoperability is a huge problem, and that problem is getting exponentially worse as data is now flowing in from so many sources, such as the new reams of patient-generated data.  E.g., to whom should I send data from my fitness tracker: my primary care doctor, my cardiologist, my health insurer, maybe even my employer?  All of them?

That's not going to work.

We're already beginning to see blockchain show up more in health care.  For example, Gem just announced Gem Health, As they say: "We need a modern infrastructure that unlocks new channels for services to connect, while balancing the need for strong data privacy and security.  Blockchain technology is that infrastructure."

Gem's vision is as follows:
We imagine a future where everyone holds the keys to their healthcare passport, bridging patient care among multiple providers and across borders.  We imagine hospitals hiring more doctors and nurses on a budget recovered from wasted reconciliation expenses.  We imagine labs, wearables, shopping lists, and healthcare apps working together to inform a healthier population.  
Philips is onboard, with the Philips Blockchain Lab joining the Gem Health network.

It's not going to be easy.  Health care has had a hard time agreeing to things like ICD-10 or HL7, much less interoperability standards.  In CIO, Peter B. Nichol points out the need for foundational protocols, such as the Linux Foundation's Hyperledger project is working on  If we thought getting providers to use EHRs was hard, picture trying to get the health care industry onto a entirely new technology platform like blockchain.

True to form, the "HIT standards mandarins" are already showing resistance to blockchain.

However, Mr. Nichol also enumerates a number of companies already jumping on the blockchain bandwagon for healthcare uses, including not just Gem but also Tierion, Factom, HealthNautica, and Guardtime.  It is something that any organization involved in health care can ignore only at their own risk.  It wasn't that long ago that Linux looked like an idealistic approach that major companies would never use, and we've seen how that has played out.

Look, I'm no technology seer.  I don't know if blockchain is going to totally revamp how we store and update data, as its proponents claim.  What I do know is that, when it comes to health care, our current approaches are not getting us to the interoperability that we need, or are doing so only at glacial speeds, and that they allow our electronic data to be increasingly vulnerable.  

If not blockchain, what?

Monday, April 18, 2016

Say Goodbye to Your Smartwatch

Just because Steve Wozniak takes a shot at Apple doesn't mean he's wrong.

Woz recently declared that the current generation of wearables, including the Apple Watch, are "not a compelling purchase."  He says that his Apple Watch is "an expense that has brought me a few extra niceties in my life," but generally is frustrated that wearables don't have enough computing power and are mostly still dependent on a linked smartphone for many of their functions.

He's not alone in his skepticism.  A trio of analysts from Pacific Coast Securities see trouble ahead for many wearable manufacturers, as "value creation shifts away from the thing itself, while the associated ecosystem, software and/or service tend to deliver the real intelligence that the things provide."  In their view, data collection is quickly becoming commoditized; e.g., it's not just enough to track steps, but it is what is done with that information that will be increasingly important.

Last summer Fitbit stock reached a 52 week high of $52; now it is trading around $17.  Looks like someone is worried.

It's not that these analysts are bearish on wearables generally; they just believe that: "consumers appear to prefer a simplified and relatively inconspicuous version of technology that collects information while leaving complex interactions to devices that are better suited for the task,"  I'm not sure Woz would agree.

Christopher Mims of The Wall Street Journal is also moving away from multi-purpose wearables to ones that have more specialized purposes.  His realization was that:
Unlike smartphones, which are pretty much the same, there is no such thing as a one-size-fits-all wearable, and there may never be. That is because we are using them to address different wants and desires.
He cites single purpose wearables like Nex Band, which bills itself as "the hackable modular smartband," and Spire, whose forte is helping manage stress by closely monitoring respiration, as examples of wearables that are "solutions to particular problems, rather than objects for which there is already a market."  Rather than a smartwatch that tries to do everything, Mr. Mims foresees lots of different wearables, in different shapes on different parts of our bodies for different purposes.

Wareable profiled the "13 hottest wearable start-ups to look out for," and only 3 of them are based on the "traditional" wristband approach.  We've got smart clothing, jewelry, earbuds, shoes, full body sensors, even smart pills!  Many people think "smart clothing" is the future of wearables, and there certainly are no shortage of companies working on them, but I think that is selling the concept short.

Mr. Mims doesn't even want to call them "wearables," since "wearing" them itself may soon be outdated.  For example, Samsung is trying to patent a smart contact lens that could project images directly into your line of sight, or take photographs.  You control it through eye motions and blinking.  No more making fun of nerds wearing Google Glasses; you won't even know they are wearing these.
Google, of course, is doing their best to keep pace.

Or your next wearable might be an "electronic tattoo."  Your skin could act as the screen, displaying, say, your vital statistics (think of hospital patients) or maybe even cat videos from YouTube.  As one of the lead researchers behind the technology said:
What would the world be like if we had displays that could adhere to our bodies and even show our emotions or level of stress or unease? In addition to not having to carry a device with us at all times, they might enhance the way we interact with those around us or add a whole new dimension to how we communicate."
Those displays may be reading information from sensors that you're not wearing, but from ones that are inside you.  Proteus Digital Health is working on not just stick-and-peel biometric patches but also on ingestible sensors, which they believe can offer closer, more accurate monitoring.  These kinds of sensors could render many types of measuring systems obsolete, and the market for them is expected to grow to close to $700 million by 2022.

I'm surprised it isn't more.

Kleiner Perkins Caufield & Byers are big believers in the future of wearables.  They say that "battery life is by far the biggest obstacle preventing broad market adoption and retention," claiming that wearables should last "weeks and months" between charges.  In their view, in the future our devices will be:
Tiny in size, little to no energy consumed during operation, constantly processing, continuously mapping our environment and gathering data, and communicating device-to-devices.  In order to make this world a reality, we'll need almost invisible components that can scavenge energy from kinetic movement and from surrounding power sources such as wireless signals.
Just as personal computers shook up our notion of what a computer was, and just as smartphones and tablets shook up our notion of what a personal computer was, wearables stand to revamp our thinking again.  Microsoft is said to be getting ready for a "world without keyboard and mouse," and we're all going to have to get used to a world where computing is so ubiquitous that we aren't even aware of the devices that are doing the computing.

We've gotten our head around Fitbit and Apple Watch, but that's just scratching the surface of what technology already allows, or soon will allow us to do, in bringing computing more into our lives -- and hopefully bringing more value to us.   It will be like jumping from a 1980's IBM PC to an iPad.

It's pretty exciting.

Tuesday, April 12, 2016

How Bot That!

About a year and a half ago I wrote I Hate Apps, expressing my concerns that apps had outlived their usefulness due to how they are cluttering up our devices, and found I wasn't alone in this attitude.  Now Facebook is doing something about it, with their vision that they can use "bots" within their Messenger app to eliminate the need for many standalone apps.

Indeed, as David Marcus, the head of messaging at Facebook, told Wired: "Everyone wanted websites when the web was launched.  And then everyone wanted apps.  This is the start of a new era."

Let's not get ahead of ourselves.  Not yet.

What is a "bot," anyway?  Here's how Kurt Wagner explained it on re/code:
A bot is software that is designed to automate the kinds of tasks you would usually do on your own, like making a dinner reservation, adding an appointment to your calendar or fetching and displaying information. The increasingly common form of bots, chatbots, simulate conversation. They often live inside messaging apps — or are at least designed to look that way — and it should feel like you’re chatting back and forth as you would with a human.
At its annual F8 developers conference, Facebook just announced that it was opening up Messenger so that other companies could build bots that would use the service to connect with, and hopefully assist, Messenger's users.  They even have a "Bot Engine" to allow bots to be much more powerful, increasing the range of tasks they can assist with.

The funny thing is, despite all the apps most people have downloaded, we typically only use a handful of them regularly.  Facebook wants to make sure it stays one of those.  They figure, why go to another app if Messenger can do the same work for you?

They already have partnerships with some 25 companies, including CNN, the Wall Street Journal, and 1-800-Flowers.  Instead of using those companies' own apps or websites, users can tell the bot what they want using Messenger, maybe even have a conversation with it to ensure it knows what you want.  Then off it goes to help you.

Mr. Marcus told The New York Times, "We're conversational creatures.  That's the way our brain functions.  That's the way we're wired.  As a result, it's probably the most natural interface there is."

That comes as no surprise to, say Amazon, which is building a whole vocal ecosystem around Alexa, or to Microsoft, whose Cortana is voice-activated and whose recent foray into bot -- Tay -- went badly awry (some say needlessly).  .

Other messaging services like WeChat already allow users to buy tickets or play the lottery, while Kik has a "bot store" that services much like an app store.  Facebook doesn't want to follow Kik's example.  As Mr. Marcus told Wired: "It's the wrong way to think about it.  It's not as if you need to go buy something or download something; the interaction is much more seamless."

This is what should have Apple and Google worried, according to Elizabeth Dwoskin of The Washington Post: both companies have made their app stores central parts of their ecosystem, so if bots replace apps, many of their current advantages dissipate.  As one venture capitalist told her, "Everyone who didn’t win the app platform wars will get another chance with bots."

Hello, Amazon, Facebook, and Microsoft.

Just as I don't want to have to figure out which app is best, download it, and then try to use it, I don't want to have to do the same with bots.  I wouldn't mind, though, telling Messenger or another service what I'm looking for, and letting the bots battle it out about which one can best help me.

I could think of a number of ways bots could help people with their health care, such as bots that can assist with the following kinds of requests:

  • Can you check my calendar and schedule an appointment with Dr. Z?  Mornings are better.
  • Can you find me a primary care doctor within three miles, who is in my insurance network, accepts new patients, and has good consumer ratings?
  • Can you check my claims, find out how much I'm supposed to owe Dr. Y, and issue a payment?
  • I'm afraid I might have high cholesterol.  Can you tell me -- preferably in a video -- what it is, how to find out if I do have it and what I'm supposed to do about it, then find me a good doctor for it?
  • Can you remind me when my next preventive service is, and schedule it?
  • I need to get an MRI -- can you find out which imaging center has the best prices for my insurance plan?
  • My father needs a heart transplant -- can you find out which programs in the country do the most and have the best outcomes?
  • Can you alert me whenever any drug on my med list goes off my health plan's formulary, or changes tiers?  Then tell me what drug I could ask my doctor to switch me to in order to minimize my copay. 
Pretty much all of these are things I could do myself, using the web and/or various apps.  It just could take a long time and be a lot of work.  I'd love to set a bot off the task and let them do the work, presumably a lot faster than I could.

Many health care organizations are patting themselves on their backs just because they have some kind of an app, even though it might be fairly rudimentary.  If they are not careful, they may find themselves once again behind on the technology curve as other industries move more rapidly towards bots.

I've written before about AI-driven personal health assistants, maybe even AI-driven virtual physicians, and I still firmly believe we'll get there someday.  Maybe bots can be the next step towards them.

Hey, if this is what it takes to make me use, and like, Messenger, that'd be OK.

Wednesday, April 6, 2016

Print Me a New Arm, Please

I just realized that I hadn't ever really written about two hot trends in health care: bionics and 3D printing.  I better get to it before they become mainstream, or are superseded by something even newer.

Let start with bionics.  According to Merriam-Webster, bionic means "having normal biological capability or performance enhanced by or as if by electronic or electromechancial devices."  Bionics is the science of this.   If you are old enough to remember the 1970's TV series "The Six Million Dollar Man" (or its spin-off, "The Bionic Woman"), bionic implants for their arms and legs gave the title characters greater strength and speed, as well as a bionic eye for Steve Austin and bionic ear for Jamie Summers.  Since they were, essentially, prototypes, six million dollars probably was a pretty good deal, even in 1970's dollars.

Flash forward to 2016.  Bionics is estimated to be an $8b market (2014), with a projected 13.2% CAGR that would bring it to $20.5b by 2020.   The artificial kidney market is said to account for some 60% of the market, with livers expected to be the fastest growing segment over the period.

Here are some other examples:
  • We've already got bionic arms, including ones with bionic hands that are controlled by the person's thoughts.  Similarly, we've already got brain-controlled bionic legs,  
  • Bionic eyes are giving sight back to some people who had lost it (even if they can't -- yet -- shoot laser beams like Steve Austin's).  We're getting to the point where we can collect visual information and send it in code the brain understands.  
  • All us aging boomers should be looking forward to bionic joints, either supplementing or replacing the ones we were born with.
  • A bionic spine is already being tested in humans, allowing paralyzed patients to walk and move their arms.
  • We're getting closer to a bionic pancreas, with companies like Beta Bionics posed to revolutionize treatment of diabetes.
None of this is cheap, of course, especially since much of it is still considered to be in the experimental stage.  A bionic arm might cost $50,000- $60,000, which is a lot more expensive than a conventional prosthesis.  Artificial kidneys just need to beat the cost of dialysis or kidney transplants to look economical.  I'll bet the medical-industrial complex is practically salivating at the prospect of us all wanting to upgrade and supersize our own equipment.

Which leads us to 3D printing.

3D printing, also known as additive manufacturing, takes a digital file and then, layer-by-layer, prints out a 3D version.  It could be a toy, a shoe -- or a body part.  3D is estimated to currently be a $4.1b market, and grow to $16.2b by 2020, with the healthcare portion alone accounting for some $3.89b of that in 2022.  Small change, perhaps, relative to the overall manufacturing and/or healthcare markets, but the potential is huge.

Newsweek recently ran an article How 3D Printing Will Make Manufacturing in America Great Again.  It may render obsolete the huge manufacturing plants -- increasingly located overseas -- in lieu of smaller plants that can make anything, anywhere.  As one 3D printing company CEO told them: "Why would you put a thousand machines in one place when you can put one machine in  thousand places?"

3D printers have been somewhat slow, but their speed is getting much, much faster.   A company named Carbon is among the fastest, and their CEO thinks even the term "printing" may be outdated.  As he told Bloomberg,  "We don't print.  We grow."  

Health care is already benefiting.  For example, a 3D printing company named Open Bionics says it can make a custom-fitted forearm and hand within two days, for about $3,000 -- versus traditional prosthetics that could cost thirty times that and take months to build.  As their founder told BBC: "We have a device at the lower-end of the pricing scale and upper end of functionality,  At the same time it is very lightweight and it can be customized for each person."

Another company, Youbionic, is taking preorders for a 3D bionic hand that run less than $1,500. and is fully functional.  As they said: :We have evaluated several choices but we have no doubt that 3D printing is what we believe to [be the] best manufacturing technology projects such as ours."

We've even got 3D pills, which raises the potential not just of reducing the cost of manufacturing drugs but raising the possibility that you could print out your own prescriptions at home.  Download the digital file, load it on your home 3D printer, and, bingo, print out your own meds.  Aprecia Pharmaceuticals has an exclusive license for an MIT 3D process for printing pills, but they won't be the only ones to go down this route (although not everyone is expecting a rush to it).

Lest anyone get nervous about the prospect of having 3D printed plastic replacing body parts, researchers are now literally able to 3D print living tissue and organs, including muscles, bone and ear tissue, even actual ears.  Bioprinting is the process of printing layers of living cells, and with new devices like a biopen surgeons can, for example, "draw" new cells directly onto bone, such as when repairing cartilage.  Researchers in Australia are also bioprinting gum and bone structures for dental repairs

Forget transplants, forget implants, forget dental fillings or crowns-- just draw desired structures directly where needed.

Bionics and 3D printing are pretty cool, and will impact health care in ways we do not yet fully appreciate.   Bionics are getting so good that people like Hugh Herr are already raising the question of whether some people might amputate their own limbs to get bionic ones.

After all, back in 2012 some athletes complained that Oscar Pistorius had an unfair advantage with his prosthetic legs, so imagine how they'd feel if a runner got bionic legs.  By the same token, why do laser eye surgery to correct your vision if you could just get a newer, better bionic eye?  It's hard to say where bionic options could lead, especially as they get both better and cheaper.

Of course, once we have nanotechnology and/or gene therapy to fix problems at a cellular level, even bionics and 3D printing may become outdated.  But those are topics for another day.