Uncertain Health in an Insecure World – 47

“Digital Infidel”

Traveling back from Harvard Business School (HBS), sitting (delayed) in an airport terminal skiving Wi-Fi, I probe the impressive deluge of digital health social media content at my fingertips.

My Twitter feed quickly links to a Healthcare IT News post announcing that the “third wave of digital” will be coming to healthcare first. “Liquid expectations” derived from consumers’ other digital experiences will extend to healthcare.

Accenture Interactive forecasts an era of health “living services”, attainable only through ‘wearables’ and so-called ‘nearables’. A revolutionary “new wave of transformative digital services” is coming, fueled by the ‘quantified self’ movement. Meeting such high expectations “… may be key for the healthcare industry – and something it’s not quite used to.”

Pivot to the 35,000+ member Digital Health LinkedIn group providing a stream of news and opinions about wearables, EMR’s, telemedicine, artificial intelligence, etc. for a range of wellness and illness applications. The group creates a safe space for established digital players and aspiring start-ups to tangle hands-free over this ever-expanding high tech marketplace.

Group leader Paul Sonnier (above) teases relevant topics, primes provocative questions, and intervenes when discussions get too testy. The group is all about wearable tech.

A mood-altering wearable from Thync (above) technology uses algorithms to relax you by activating parasympathetic pathways. Chilled out… or addictive?  Voice monitoring apps from the University of Michigan detect mood swings in bi-polar individuals. Mood equalizing… or altering? I've contributed to this real-time dialog, and have occasionally ‘liked’ others’ comments.

But despite strong peer oversight, some Digital Health messages feel like a marketeer’s drumbeat.

McKinsey Quarterly’s insights app alerted my iPhone to the Novartis plan for “digitizing medicine in an aging world”, and how this will drive innovations in healthcare. As the world’s population swells by >1 billion over the next decade, >50% of that increase will be people >50 years old. CEO Joseph Jiminez points to big data analytics as a critical Big Pharma capability for genomics-targeted drug development in personalized medicine, and for precision prescribing in complex chronic diseases. A 2014 Novartis-Google partnership produced a digital contact lens (below) that monitors blood sugar in diabetes, while refracting vision.

Mr. Jiminez guesstimates that 25% of global healthcare spending is wasted; digital data-feeds and predictive analytics could drive business efficiencies in response to downward pressures on R&D costs and drug prices. He predicts positive outcomes from Novartis partnerships with data & tech companies, offering a final C-suite thought, “… you get rid of everything else, and physical (sic) medicine is going to be part of that.”

Oddly enough, Novartis’ CEO did not mention any academic health center research partnerships. 

The week’s HBS executive program on Business Innovations in Global (BIG) Health Care gathered colleagues from around the world. Whether it was tech sector Cisco manager Cem from Europe or consumer-facing entrepreneur Kunal from India, in-class pitches focused on innovations in digital health platforms. But it will be the integrators who will usher in the ‘Third Phase’ of global health by financing global scale innovations in the digital sector.

And as reinforced at HBS, there is a BIG shift of the smart money towards digital health sector opportunities. Professor Michael Chu put it bluntly to the class, “… resources are reserved for the winning ideas.” While private equity firms with strong initial business positioning come and go, new industry sectors are created (and sustained) by above-average commercial returns. A recent internet of things (IoT) analytics industry sector report projected market growth from US$4.85B to US$16.35B between 2015 and 2020.

Winning!!! 

      

My delayed flight is finally boarding, as I open the New York Times.

An article on ISIS and the Lonely Young American by Rukmini Callimachi chronicles the textbook ISIS radicalization of a young woman living in rural Washington – call her Alex (above). Alex’s social media and Skype conversion ended with her Twitter witnessed recital of the Shahada, “There is no God but Allah…” Her handlers pointed out that following non-Muslim ‘kuffar’, or infidels, on social media was no longer tolerable. Alex’s family eventually intervened, and the ISIS recruiters went dark… until she reconnected months later, only to resume the chatter. Alex’s ISIS ‘friend’ had promised her family that he would cease all on-line communications…

“But I lied”, he later revealed.

Social media forecasts & figures on digital health are not lies. But this industry sector desperately wants to convert us into believing that they are telling The Truth.

As currently configured, no health system in the world is viable through 2025 – continued status quo spells disaster! In the face of such dire Future Watch scenarios, the digital health sector is doubling down on health system salvation via the ‘integration of things’.

While at HBS, financier Francois Maison-Rouge cautioned that, “If doctors find something that works for patients, they will not change to adopt a new technology.” The only viable conduit to cracking this healthcare inertia is innovation.

We in the Square desparately want to believe. But if "Video killed the radio star" in the 1970's, then will "Digital kills the clinical star" in the 2020's?

The flight attendant has unexpectedly taken away my hand held device... Didn't see that one coming!

Uncertain Health in an Insecure World – 46

“What is… Elementary?”

Jeopardy is a TV game show that requires contestants to answer in the form of a question. IBM Watson, named after former IBM president Thomas J. Watson Jr., won the show in 2011 over two humanoid champions.

Since early 2014, the IBM Watson Group has rolled out a campaign for U.S. and world market domination in the cognitive computing domain.  While its global networking has largely focused on financial services at CaixaBank in Spain and ANZ Global Wealth in Australia, the game show winning supercomputer has also been deployed to Bumrungrad International Hospital in Thailand to enhance cancer care, and to Metropolitan Health system in South Africa to tackle personalized medicine.

Overall, Watson is becoming literate in >25 countries around the world.

As I saw last week at the IBM Watson client experience center in New York’s fashionable lower Manhattan ‘Silicon Alley’ (above), the machine’s capacity to master the content of 23,000 Medline bio-medical journals gives Watson all the answers to any reasonable question that doctors and scientists might pose. Its natural language processing (NLP) skills allow it to pull in structured data and textual electronic medical record (EMR) entries, and connect these to increasingly common genomic testing data in the charts of patients receiving care in hospitals & clinics, often as part of clinical trials.

While capturing information from the medical literature and iterating it with the contents of the clinical record, Watson learns from its mistakes by using scoring agent algorithms to compute the strength of published evidence against the clinical outcomes of patients. In this way, Watson’s machine learning capabilities as a clinical decision support tool can rapidly generate high confidence diagnoses that could, in theory, save lives and improve care efficiency.

Such artificial computer “expertise” doesn’t replace the hard won clinical acumen of skilled consultants, but it can be a big boost to busy doctors caring for complex cases, working under the growing pressures of contracting resources and health system penalties for poor outcomes and long hospital stays. The chief medical information officer at Bumrungrad Hospital, Dr. James Miser, says “Watson is like having a capable and knowledgeable colleague who can review the current information that relates to my patient.”

   

Cancer care and research are very hard to separate these days; >90% of patients entering clinical trials have genomic &/or proteomic data as key qualifying criteria. Much of Watson’s current cancer capabilities were learned from the best practices and research files of New York’s Memorial Sloan Kettering Cancer Center. Lest our four-legged friends be left out, Watson has engaged the ag-vet expertise of Guelph University and LifeLearn in Canada to develop a mobile app for animal doctors to weigh treatment options on the fly, a process of veterinary care now being tested at New York’s Animal Medical Center (not yet subject to the U.S. Obamacare regulations!).

Question: Poor and less developed countries.

Answer: What is… Watson’s brave new world?

If IBM CEO Ginni Rometty (above) has something to say about, “He” definitely will be there. At the intersection of human soft intelligence and big data, poor and less developed countries can more cheaply use the cloud to interface with Watson, without the cost and risk of local site investments. The relative absence of legacy computing systems and big data platforms makes Africa a high tech desert, but bodes well for rapid Watson adoption.

As part of IBM’s Project Lucy in Africa, Rometty points to the Digital Aquifer project that will provide “… analytics on 100 million points of water across the African continent… It’s going to reinvent drought management.”

Statistics South Africa called the state of data collection on the continent, “Africa’s statistical tragedy”. Other Project Lucy initiatives include analyzing the impact of food price patterns on nutrition, and the relationship between poverty and endemic diseases. A well deployed Watson healthcare strategy could empower beleaguered health workers dealing with HIV-AIDS and drug-resistant tuberculosis in sub-Saharan Africa.

IF, and I still say IF, Watson can speed patient care and reduce deadly medical mistakes, then it will “do no harm”, and probably provide a lot of real good. While the whole Watson platform has the sniff of high tech marketing hyperbole, its mad cognitive computing skills combined with true medical care expertise and constantly refreshed cutting-edge research sourcing is hard to deny, at least out of hand.

So, as the complexity of it all continues to escalate, we in the Square will join the IBM Watson team, and give the new Big Blue a chance to bring practice changing research to global health systems that have become sloppy, expensive and frankly dangerous. 

Uncertain Health in an Insecure World - 45 

"Artificial Intelligentsia"


Next week, I’ll Waze my way into lower Manhattan to join other healthcare leaders to interface with the IBM Watson Group. IBM is investing billions in artificial intelligence (AI) for business integration. Our university recently bought the suburban New York building where Watson was developed. Watson is a cognitive computer, made famous by its successful run on the TV game show Jeopardy. Unlike other big data crunching supercomputers, Watson’s advanced natural language processing capabilities helps it bridge the masses of text information in the medical literature and EHR patient records with structured genomics data.

The Cleveland Clinic, The Mayo Clinic, MD Anderson and Memorial Sloan-Kettering Cancer Centers are working with IBM on personalized medicine approaches, but the projects are lagging. That said, Harvard medicine professor and Beth Israel Hospital Chief Information Officer John Halamka harvested 3 petabytes of big data from all the Harvard hospitals in 2011, and used his wife's personal genomics analytics to help doctors cure her breast cancer. "In today's healthcare system, it takes, on average, 20 years for an innovation from one hospital to diffuse throughout the country", worried Dr. Halamka. In his CXO-TALK (@cxotalk and http;//www.cxotalk.com), he also warned about all the big data applications being hyperbole from vendors. But at a personal care level, for his wife, big data analytics was a game changer! 

I am not a computer scientist, so I've decided to study up before meeting Watson. Of course, because all the world's computer scientists, Hadoop programmers, and algorithm diviners are populating my Twitter feed, I went directly to the source. At the IBM blog site, I read about "enabling the world: IBM Bluemix, the cloud and cool apps". Apparently, to interact with Watson, you will need to "add and angular JS interface to a Node-RED workflow...". OK... sounds easy enough, and there is a diagram (see below).

Apparently, this shows us how to quickly create a Watson user interface without installing any applications locally. Who knew? And the IBM Watson Translation app looks very happy about this... in two languages!

In healthcare, AI has proven better at the more mundane tasks of reviewing hospitals billing records, and for analyzing the insurance mix, patient demographics, physician workload & other variables affecting revenues and cash flow. As Obamacare squeezes the U.S. healthcare system hard, this kind of artificial intelligence is becoming really material!

Amid the explosion of personal Internet-of-Things little data and the streaming of unstructured medical device data, behind the scenes progress is being made by math and engineering 'quants' on tagging key data and building smarter algorithms. Quasi AI machine learning firms can easily iterate thousands of model algorithms weekly, helping Osco Pharmacy to figure out the people who bought beer when buying diapers. But only the quants know what goes into a model, and how to make sense of it.

Walmart and its 250 million weekly customers generate 40 petabytes of personal transaction data daily... "Walmart has made big data part of their DNA".  Big Pharma also collects many petabytes of personal prescription information daily from pharmacies and patients using tens of thousands of sources – trying to leverage this big data to predict populations that might be receptive to new medicines. Conversely, in their clinical trials, Big Pharma companies use big data analytics to de-risk their investments in new drugs by discerning which of these drugs are failing fast, in part in order to conserve their R&D budgets.

Computer scientists have long dreamed of building machines that can mimic our minds and preserve our memories after death. Neuroscientists are coming around to the idea that brain-to-computer data transfer is not science fiction. When Ex Machina meets the cloud, the ending may not be a happy one.

 Oxford's Future of Humanity Institute recently reported on the technical requirements for whole brain emulation, but the ethics are still dicey. The field of computational neuroscience has emerged to tackle this complexity, and biotech start-up Halcyon Molecular founded by PayPal co-founders Peter Thiel and Elon Musk is very interested in the DNA-AI interface. The U.S. DARPA-backed Brain Research  Through Advancing Innovative Neuro-technologies (BRAIN) project and the European Human Brain Project are pushing the scientific envelope. They estimate it would take 1,000 terabytes to store one brain's content (U.S. Library of Congress would take 10 terabytes to store).

What does ALL this data and computing capacity coming together really promise?

I plan to ask Watson the question that 'Joshua' was asked in War Games (1983)...

Until that meeting, we in the Square have a very human intuition that much of the future promise of AI is actually hyperbole in the present.

Uncertain Health in an Insecure World – 44

“Watchers”

“The whole world’s watching… The whole world’s watching,” was the chant from the protesters at the 1968 Democratic Convention. As the Chicago police cracked heads and enforced The Law on the streets, the muzzling of a free society and free speech was the abiding concern. Today, the unmuzzling of society and speech is the emerging problem.

Today, there are no secrets.

As was noted in post #21, an expectation of privacy is not a guarantee of same.

As the National Security Agency (N.S.A.) monitors U.S. citizens’ domestic phone calls for national security reasons, the promise of the greater good is that we’ll all be better off for that loss of our personal privacy. While the U.S. Senate debated arcane liberty-security trade-offs, The Patriot Act remained in effect under U.S. Congressional oversight, as it has since December 4, 1981 (a day that will live in infamy for civil libertarians).  Executive Order 12-333, which “exercises the necessary supervisory control to ensure compliance with the regulations”, is the epitome of Reagan era “We’re from the government and we’re here to help” intrusions (see post #6). In November 2014, The USA Freedom Act extended the “lone wolf” and “roving wiretap” provisions of The Patriot Act. Last week, the U.S. Senate rolled back these N.S.A.’s surveillance powers for the first time since 9/11.

Writing this month in Foreign Affairs on ‘The Violence of Algorithms’, Taylor Owens describes the intelligence analytics firm Palantir (http://www.palantirtech.com/) – their software is widely used for dense data visualization by the N.S.A., F.B.I., C.I.A. and other police forces (like New York).

Silicon Valley CEO PeterThiel, who previously launched PayPal, made billions on this security platform. Palantir’s company mantra is “radical transparency”, an ideal that “every employee should have the right to know everything that goes on in the company”, unless of course it’s a secret! The company’s goal is to “make a product that spans every step of analysis… from data integration all the way to end-user visualization and presentation.” For those of you who, like me, have found themselves trying to get out of The City using Waze, it’s obvious that this widely used traffic avoidance app (purchased by Google in 2013) is the non-combatant version of Palantir.

Recently, the so-called Dark Net (or Deep Web) internet has been uncloaked from the shadows. Its origins are traced to The Onion Router (Tor) encryption software developed at the U.S. Naval Research Laboratory released in 2002. Tor trafficking conceals user location & identity, permitting internet communication anonymity.   Tor usage exploded in 2013 when the N.S.A.’s domestic surveillance program was revealed.

An emerging Dark Net figure is Silk Road founder Ross Ulbricht, who was sentenced to life in prison in March 2015 for drug dealing and conspiracy charges tied to his online marketplace. He collected $18 million in bitcoin credits – the untraceable crypto-currency of the Dark Net. In 2014, the CEO of the BitInstant bitcoin exchange, Charlie Shrem, was charged with illegal money laundering for online drug purchasing by Silk Road users. Dark Net devotees put forth a libertarian philosophy in defense of this activity, saying it supports individual freedom & protects privacy. But libertarian fears about state encroachment into personal affairs & freedom of expression do not excuse how this technology has morphed from good intents to evil actions. Over 50,000 Tor browser websites are notoriously used for child pornography, drug dealing and on-line bullying called “trolling”.

Not all Deep Web use has such dark connotations. Besieged human rights activists, journalists and citizens in high risk countries around the globe have stayed connected using the Tor network. Chinese dissidents, Arab Spring-ers, Iranians, Syrians and victims of domestic abuse all have been able to anonymously seek online support.

The Dark Net author Jamie Bartlett even suggested that Anonymous wage an on-line war against ISIS recruiters on the Dark Net!

All of these technology pushes and pullbacks reflect a highly dynamic system that is machine learning in response to a wide array of social media trends and societal ques.

Like you, we in the Square have long operated under what, in retrospect, is a false sense of security.

Can't you feel the Watchers cracking us on the head?

Uncertain Health in an Insecure World – 43

“Inter-not of Things”

Like many of you, my Twitter feed is deluged with tweets from very sophisticated denizens of the #bigdata and #analytics domains. For those of us who cannot distinguish a CEO from an SEO, it’s hard to keep up with the related buzz feed.

BTW, a SEO is a search engine optimizer.

And those analytics folks are way smart. They can take ALL that big data and fit it into mathematical models – algorithms – that can predict something useful about what is likely about to happen before it actually does. 

An excellent recent article on the topic by Jennifer Bresnick in HealthITAnalytics neatly defined big data analytics as “joining of two or more previously disparate sources of information, structured in such a way that insights can be drawn from the comparison or examination of the new expanded data set.” I really like when people make complex things simple!

In healthcare, ubiquitous electronic health records (EHR’s) contain oodles (a technical term still not owned by Google) of structured and unstructured big data. Some of this structured personal health information is routinely entered into the EHR by healthcare workers – patient demographics, vital signs, drug allergies, lab test results, etc.

BTW, it is my casual observation that there are still more healthcare workers than sex workers on Twitter, but one wonders how long that will last.

Other more complex structured data is laboriously captured and entered by busy physicians who must populate required clinical data fields and reconcile medications between patient visits or after clinic hours. All this is then washed over by a tsunami of unstructured data streaming in from bedside monitors and other medical devices in and around the environment of care.

So far, the healthcare industry has found it very difficult to bring disparate big data sets together to improve business practices or patient outcomes. The same 5,000 U.S. hospitals that generate over US$850 billion in annual revenues has yet to crack the big data analytics ‘nut’. Why is that? Is it EHR vendor incompetence? Is it a paucity of highly qualified personnel in this health IT domain? Is it the failure of policy-makers to align population health goals with early adoption? Of course it is!!

But we can now add the Internet of Things (#IoT) to the mix. The explosion of smart phones and #wearables make up a “nebulous network of collection nodes”. These things, almost any device with a charger or batteries, are Internet-connected to a larger network. Once these uniquely identified things start storing information in The Cloud, supported by the free apps that allow for real-time consumer interface, then you have the IoT! Personalization of these things is possible… Hell, it’s encouraged!

BTW, a Premium LindedIn package quickly reveals most of your personal attributes and professional talents to other world-wide invitees… How far off is it until the IoT knows your pulse?

   

The global exemplar of the IoT shaping healthy behavior changes may be found in a place as unexpected as post-Apartheid South Africa, where a company called Discovery has been innovating on the healthcare insurance pitch. CEO Adrian Goree outlined this novel platform in the current McKinsey Quarterly. Discovery’s Vitality™ program logs in Next+ and Fitbit data on 70,000 member fitness workouts per day via mobile apps and smartphones, rewarding their customers with 25% discounts on a total of $100 million of healthy foods. Vitality™ and other nudge incentives have eased Discovery’s access to global healthcare insurance markets in China, Europe, the U.S., etc.

Until better natural language processing algorithms are developed to digitize medical chart wording, the richly nuanced histories told by patients to their doctors, then entered into the EHR, are hard to use. As such, the phenotyping of patients’ individual disease manifestations is much more difficult than is the genotyping of their DNA. The Salt Lake City-based John Huntsman Cancer Institute – steward of the Utah Population Database (UPDB) – has asserted the importance of population genotyping to medicine. But one without the other is not nearly as powerful as both.

BTW, did you know that the popular Ancestry.com genealogy website is owned by the Church of the Latter Day Saints – the Mormons? So, the UPDB in SLC contains mostly LDS info... That makes me want to sing & dance!

My oft-favorited Twitter follower and newly LinkedIn colleague is Dr. Morten Middelfart, the CIO at Genomic Expression. Morten is flat-out brilliant, as his recent Oxford University lecture on “the magic” of big data analytics for solving chronic disease mysteries revealed on Twitter bit.Iy/1CHOIEp His Denmark-based firm has clearly demonstrated how Cloud-based RNA seq analysis and related algorithms can accurately model frequently ineffective therapies (see below).

The names that we give to the social media tools that monitor our collective information – the data-ome – are purposefully designed to be comfy, often bordering on the banal. Twitter… Facebook… the Internet of Things…

And the advent of peaceful & serene Cloud and open-architecture computing by the cute yellow Hadoop elephant should not be misunderstood as either light and fluffy, or dumb like Dumbo. 

To a hammer, everything is a nail!

What remains unproven is whether the power of Internet of Things tools can be harnessed, or not. 

We in the Square are not yet persuaded that IoT data are truly powerful, because “… big data is only as smart as those who generate it”.