IBM Watson Health’s stumble and possible fall

July 21, 2021 | by

This Editor hadn’t thought about or seen news about IBM Watson Health in over a year…and likely, neither did you. Granted, our minds have been Otherwise Engaged, but for the company that was supposed to dominate AI and health analytics, it’s notable that TTA’s last two articles mentioning Watson Health was 25 April 2019, on a report that its Drug Discovery unit was being cut back as the latest in a series of executive cutbacks and lawsuits (MD Anderson on a failed oncology initiative), and 14 Feb 2020 on 3M’s lawsuit on unauthorized use of their software.

The New York Times in an investigative piece (may be paywalled or require signup for limited access), brings us up to date on what is happening at IBM Watson, and it’s not bright for Watson Health. IBM, like so many other companies, badly underestimated the sheer screaming complexity of health data. Their executives believed they could translate the big win on the “Jeopardy!” game show in 2011, based on brute computing power, into mastery of healthcare data and translation into massive predictive models. The CEO at the time called it their ‘moon shot’. Big thinkers such as Clayton Christensen chimed in. IBM managers sang its praises to all in healthcare who would listen. This Editor, on a gig at a major health plan in NJ that was ‘thinking big’ at the time and used IBM consultants extensively, in 2012 was able to bring in speakers from Watson for an internal meeting.

But we haven’t been on the moon since 1972 (though probes have visited Mars). Since the big push in 2011-12, it’s been one stumble after another. According to the Times:

  • The bar was set much too high with oncology. Watson researchers knew early on in their research at the University of North Carolina School of Medicine that their genetic data was filled with gaps, complexity, and messiness. The experience was similar with Memorial Sloan-Kettering Cancer Center. The products growing out of the UNC and MSKCC research, Watson for Genomics and Watson for Oncology, were discontinued last year. These were in addition to the MD Anderson Cancer Center initiative, Oncology Expert Advisor for treatment recommendations, that was kicked to the curb [TTA 22 Feb 2017] after $62 million spent. At the same time, IBM’s CEO was proudly announcing at HIMSS17 that they were betting the company on multiple new initiatives. 
  • Watson Health, formed in 2015, bought leading data analytics companies and then didn’t know what to do with them. TTA noted in August 2018 that Phytel, Explorys, and Truven Health Analytics were acquired as market leaders with significant books of business–and then shrank after being ‘bluewashed’. HISTalk, in its review of the Times article, noted that along with Merge Healthcare, IBM spent $4 billion for these companies. IBM’s difficulties in crunching real doctor and physical data were well known in 2018 with revealing articles in IEEE Spectrum and Der Spiegel. 

Six years later, Watson Health has been drastically pared back and reportedly is up for sale. Smaller, nimbler companies have taken over cloud computing and data analytics with AI and machine learning solutions that broke problems down into manageable chunks and business niches.

What’s recoverable from Watson? Basic, crunchy AI. Watson does natural language processing very well, as well as or better than Amazon, Google, and Microsoft. Watson Assistant is used by payers like Anthem to automate customer inquiries. Hardly a moonshot or even clinical decision support. For business, Watson applications automate basic tasks in ‘dishwashing’ areas such as accounting, payments, technology operations, marketing, and customer service. The bottom line is not good for IBM; both areas bring in a reported $1 billion per year but Watson continues to lose money. 

Blockchains, EHRs, roadblocks and baby steps

March 12, 2018 | by | 3 Comments

TTA founder and former editor Steve Hards crawls out of his retirement tent to squint at the misty landscape of blockchain technology.

In a recent dream I was observing an auditorium full of people chanting “Blockchain! Blockchain! Blockchain!” and yes, mantra-like, blockchain is now popping up all the time in health technology articles and presentations.

It has taken a while to get to this stage. It was January 2016 when Editor-in-Chief Donna first mentioned blockchain. Since then there appears to have been more talk than action.

A year ago, in February 2017, health IT guru Brian Ahier was able to say in a comment here “Blockchain of course, is going to sneak up on a lot of people…”

Where we have seen developments occurring is in the trickle of ‘coins’ or ‘tokens’ in health-related Initial Coin Offerings (ICOs) of dubious investment worthiness. I may rant about those in a follow-up article if anyone is interested. (Let me know in a comment.)

The terminology is still in its ‘shakedown phase’ (see this great terminology rant) and, because of the publicity around Bitcoin, which is on a blockchain, the distinction between blockchains and distributed ledger databases is blurred. There are technical differences: blockchains are a sub-set of distributed ledgers (Wikipedia), which is the term I’ll generally use in this article.

Distributed ledgers and EHRs

What are the implications of distributed ledgers for the biggest databases in healthcare, electronic health records (EHRs)?

The two principal characteristics that differentiate distributed ledgers from the databases with which we are familiar are that they are more robust and, potentially, more private. Some even claim to be quantum computing hack proof although we will have to wait for hackers with quantum computers to test that.

Traditional databases are formed from one large or several linked entities that have a centralised control from where performance, data integrity and security are monitored and managed. There are human and technological factors that introduce weaknesses to all such systems, as the number of data breaches reported here over the years testify.

(more…)

DNA ‘Snapshot’ facial modeling–and predicting future Alzheimer’s risk

May 31, 2017 | by
[grow_thumb image=”https://telecareaware.com/wp-content/uploads/2017/05/SNPSHT-Example-1-1024×972.jpg” thumb_width=”150″ /]It sounds like something from an episode of ‘Law & Order’ (US or UK), but extracting facial appearance and ancestry from a forensic DNA sample isn’t fiction anymore. Parabon NanoLabs was funded by the Defense Threat Reduction Agency (DTRA) to develop Snapshot originally to dismantle improvised explosive device networks in Iraq and Afghanistan. The methodology was then transferred to DNA analysis. Parabon uses data mining and advanced machine learning to predict how the single nucleotide polymorphisms of the genome will make someone appear. This appearance profiling includes eye color, skin color, hair color, face morphology, and detailed biogeographic ancestry (see left above). The forensic art alone can age up or down the subject, adding or subtracting glasses and facial hair. These factors have successfully focused investigations for over 80 law enforcement agencies. According to Armed with Science, Parabon is now transferring the technology to predict an individual’s lifetime risk of Alzheimer’s–certainly a revolutionary use in healthcare technology.

A breakthrough wearable? Sweat analysis for cystic fibrosis and diabetes diagnosis.

April 19, 2017 | by
[grow_thumb image=”https://telecareaware.com/wp-content/uploads/2017/04/Sweat-Sensor-Stanford.jpg” thumb_width=”175″ /]Researchers at Stanford University School of Medicine and University of California-Berkeley have developed a wristband equipped with a sensor that can capture and analyze perspiration. The design stimulates the production of sweat, with the embedded sensors and microprocessors detecting the presence of different molecules and ions based on their electrical signals. In the abstract’s words, this is an “electrochemically enhanced iontophoresis interface, integrated in a wearable sweat analysis platform.” The wearable was tested in two separate studies for detecting a key indicator for cystic fibrosis (CF)–a high level of chloride ions–and in comparing levels of glucose in sweat to blood glucose for diabetes. The data is transmitted via smartphone to a server that analyzes the results in real time.

The potential for this wearable is considerable. First, for CF, it changes a 70-year-old protocol–that sweat is stimulated and collected in a 30-minute procedure, then sent to an outside lab to be analyzed with the usual delay. Children being screened for CF have trouble sitting still for the lengthy test. The second is that the test can be done anywhere with minimal training, making it suitable for underserved communities and developing areas of the world. The third is in CF drug development. CF genetics have multiple mutations, limiting drug usefulness. A test such of this in real time could speed drug clinical trials and human response.

The glucose testing was preliminary in comparing the glucose in sweat with standard blood glucose levels, but also proved that the platform could be used for other perspiration constituents, such as sodium and lactate. The ultimate intent of the researchers is to incorporate the technology into a smartwatch for continuous monitoring, but they recognize two challenges: reproducibility, to see whether measurements are consistent, and mapping all the constituents of sweat.

The report was published on 17 April in Proceedings of the National Academy of Sciences of the United States of America (PNAS). Abstract and full report (PDF, 6 pages). Stanford Medicine News Center

Eight TECS expected to change health and care

January 26, 2016 | by

[grow_thumb image=”https://telecareaware.com/wp-content/uploads/2016/01/8-technologies-8-connected-community.jpg” thumb_width=”175″ /]The King’s Fund is still bullish on the transformative capabilities of technology-enabled care services for health (even if others are not, see following article). This article (which almost passed this Editor by this month) highlights eight areas which have the greatest potential. Some are expected–but at least two are surprises. You be the judge!

  1. Smartphones: apps, as hubs/hub replacements, and research transmitters (voluntary but also involuntary?)
  2. At-home and portable diagnostics; smart assistive technology
  3. Smart or implantable drug delivery
  4. Digital therapeutics/interventions; cognitive behavioral therapy; lifestyle interventions
  5. Genome sequencing
  6. Machine learning (computers changing based on new data, spotting pattern) in big datasets (Surprise #1)
  7. Blockchain, the tech behind bitcoin; decentralised databases, secured using encryption, that keep an authoritative record of how data is created and changed over time, to bring together decentralized health records. (Surprise #2)
  8. The connected community; P2P support networks and research communities

The King’s Fund’s publications 1 Jan

Weekend must read: The Death of Patient Zero

July 31, 2015 | by
[grow_thumb image=”https://telecareaware.com/wp-content/uploads/2015/07/landscape-1438023958-esq080115stephanielee001-hope.jpg” thumb_width=”150″ /]The story of one woman with advanced cancer–Stephanie Lee–as doctors and researchers at Mount Sinai NYC race to save her with genomics-driven personalized medicine. We see its limitations, along with the limitations of conventional medicine and the problems of the stateside military medical system–Mrs Lee’s husband was killed in combat in Iraq in 2005. What was unlimited was the courage of her family, her friends and her medical advocates, especially one of those Mount Sinai genomicists, Eric Schadt, an “evangelical Christian turned mathematician turned biologist turned genomicist who had become one of the evangelical forces behind the “Big Data” revolution” and Dr Dennis Charney, the head of Mount Sinai’s Icahn School of Medicine who has made a home for gene sequencing research there. Tom Junod writes about Patient Zero in Esquire –including why she was given that name.  Photo–Esquire

Do startups truly threaten the ‘healthcare establishment’?

May 19, 2015 | by

Or are successful startups fitting into their game? Chris Seper in MedCityNews paints the picture of one side of a quandary. The ‘healthcare establishment’ fundamentally and to its detriment does not understand and is threatened by the startup and innovation process. A startup may begin with an idea which is, in his words, ‘almost always flawed, sometimes deeply’. If the founders are smart, they will test their ideas, validate them and change them appropriately. If not, they will fail. But it is easier for the Establishment to point at the most egregious of the bad ideas and use them to rationalize the status quo.

But being congenital contrarians, we paint the house on the other side of the street. Has the Establishment caught up with–or in some cases, co-opted startups, making them and their funders ‘do their diligence’ and be more cautious before emerging? This Editor would argue yes, and largely for the better.

**The ‘Wild West’ days are over. A few years ago, a truly bad or deeply flawed health tech idea or could easily find funding, because it was all blank slate, new and ‘transformative’.The sexiest hooks were Quantified Self, sleep, employer health incentives, interactive coaching, genomics, app prescribing and (last) wearables. A lot of founders imagined themselves as the Steve Jobs of Healthcare, down to the black turtleneck. Now there is a history of success and failure. The railroads reached the dusty frontier towns.

**There’s now a ‘Startup Establishment’. National accelerators (more…)

ELabNYC Pitch Day 2015

April 4, 2015 | by

Thursday 3 April, Microsoft’s NY Technology Center, Times Square NYC

[grow_thumb image=”https://telecareaware.com/wp-content/uploads/2015/04/Elab.png” thumb_width=”100″ /]The third annual Pitch Day for the now 20 startup/early-stage life science, biotech and healthcare technology companies in the ELabNYC (Entrepreneurship Lab Bio and Health Tech NYC) is a culmination of their year-long program participation in this NY Economic Development Corporation (NYCEDC)-supported program. The entrepreneurs in the ELabNYC program primarily come from from the doctoral and post-doc programs from New York’s many universities, from CUNY to Columbia, from many parts of the world, and most have experience within the city’s multitude of major health research institutions from The Bronx to Brooklyn. New York is also a center of funding for life science and health tech ventures; it’s #2 with NIH awards totaling $1.4 billion. For the past few years, NYEDC has also supported these companies with finding access to capital, specialized space (e.g. wet labs such as the million square feet at Alexandria Center alone, plus Harlem Biospace and SUNY Downstate in Brooklyn) and partnerships with major companies such as Celgene, Eli Lilly, Pfizer and GE Ventures.

This Editor will concentrate on health tech companies–eight, up from five last year [TTA 17 Apr 14]. Each company pitched for five minutes on its concept, its current state of advancement (including pilots/customers), its team and a funding timeline. It was a very different mix from last year’s class, which focused on compliance, diagnosis, dementia and concussion. These companies focused on niches which are either not being served well or to substantially reduce costs. Nearly half the entrepreneurs were women, a substantially greater number than one usually sees in the biotech/health tech area. Short impressions on our eight, with links to their Executive Summaries on the 2014-15 ‘class page’: (more…)

Dr Topol’s prescription for The Future of Medicine, analyzed

January 12, 2015 | by

The Future of Medicine Is in Your Smartphone sounds like a preface to his latest book, ‘The Patient Will See You Now’, but it is quite consistent with Dr Topol’s talks of late [TTA 5 Dec]. The article is at once optimistic–yes, we love the picture–yet somewhat unreal. When we walk around and kick the tires…

First, it flies in the face of the increasing control of healthcare providers by government as to outcomes and the shift for good or ill to ‘outcomes-based medicine’. Second, ‘doctorless patients’ may need fewer services, not more, and why should these individuals, who represent the high-info elite at least initially, be penalized by having to pay the extremely high premiums dictated by government-approved health insurance (in the US, ACA-compliant insurance a/k/a Obamacare)–or face the US tax penalties for not enrolling in same? Third, those liberating mass market smartwatches and fitness trackers aren’t clinical quality yet–fine directionally, but real clinical diagnosis (more…)

Data mining health records: the good, bad and ugly

October 2, 2014 | by

Take your time this weekend and read this article from the Washington Post on the ‘brave new world’ of data mining health records. While those with experience analyzing real-world health data snicker at Larry Page of Google’s inflated claims of ‘saving 10,000 lives in the first year’ if only he could get his hands on that identified data (of course, then there’s the opportunity to make $£€¥, which is what Larry and Sergey are really interested in–count your Editor as a cynic!), the Health Data Analytics Express rolls on. The promise lies in more precision in treatment areas such as brain tumor radiology where sizing is critical (BraTumIA) and individualized genomics for disease. Yet the author does not touch on healthcare decision support systems best exemplified by IBM Watson, (more…)

23andMe as Rorschach test

December 14, 2013 | by
[grow_thumb image=”https://telecareaware.com/wp-content/uploads/2013/12/rorschach1.jpg” thumb_width=”125″ /]Just like the good doctor’s ink blot, there’s a lot of ‘reading into’ the travails of genomic ‘spit test’ 23andMe. Blogger PF Anderson, the Emerging Technologies Librarian for the University of Michigan Health Sciences Libraries, has collected them in pithy quotes and citations. For your weekend or holiday exploration: Collecting Thoughts on the FDA vs. 23andMe

Snowden and digital health – the FT finds a worrying connection

August 2, 2013 | by

The FT’s excellent journalist Gillian Tett has written a thought-provoking article on how increasing privacy concerns brought about by recent cyber surveillance revelations are threatening the ability to use ‘big data’ to connect specific genomic features with individual health conditions.  This in turn is threatening the ability to find and improve treatments and cures for many ailments.  All is not lost though; she describes a number of possible solutions, most notably a “people’s movement”.

Well worth a read even though it fits more under a wider definition of Digital Health than is conventional on TTA.

(Holiday) Weekend reading: McKinsey’s guide to 12 disruptive technologies

May 23, 2013 | by

The McKinsey & Company consultants have compiled two lengthy PDFs (one long executive summary and a very long full study), plus a podcast by one of their researchers, on what they see are 12 core disruptors which will be familiar to most of our readers. None are labeled ‘healthcare’ but seven of the 12 fit right into any tech in the field: mobile internet, the ‘internet of things’, advanced robotics, automation of knowledge work, cloud computing, next-gen genomics and 3D printing. Disruptive technologies: Advances that will transform life, business, and the global economy (downloads in article)