The confusion within TEC/telehealth between machine learning and AI-powered systems

Defining AI and machine learning terminology isn’t academic, but can influence your business. In reading a straightforward interview about the CarePredict wearable sensor for behavioral modeling and monitoring in an AI-titled publication, this Editor realized that AI–artificial intelligence–as a descriptor is creeping into all sorts of predictive systems which are actually based on machine learning. As TTA has written about previously [TTA 21 Aug], there are many considerations around AI, including the quality of the data being fed into the system, the control over the systems, and the ability to judge the output. Using the AI term sounds so much more ‘techie’–but it’s not accurate.

Artificial intelligence is defined as the broader application of machines being able to carry out tasks in a ‘smart’ way. Machine learning is tactical. It’s an application that assumes that we give the machine access to data and let the machine ‘learn’ on its own. Neural networks in computer design have made this possible. “Essentially it works on a system of probability – based on data fed to it, it is able to make statements, decisions or predictions with a degree of certainty.”, as stated in this Forbes article by Bernard Marr.

CarePredict has been incorporating many aspects of machine learning, particularly in its interface with the wrist-worn wearable and its interaction with sensors in a residence. It gathers more over time than older systems like QuietCare (this Editor was marketing head) and with more data, CarePredict does more and progressed beyond the relatively simple algorithms that created baselines in QuietCare. They now claim effective fall detection, patterns of grooming and feeding, and environment. (Disclosure: this Editor did freelance writing for the company in 2017)

In wishing CEO Satish Movva much success, this Editor believes that using AI to describe his system should be used cautiously. It makes it sound more complicated than it is to a primarily non-techie, senior community administrative and clinical audience. Say what you do in plain language, and you won’t go wrong. AI for Healthcare: Interview with Satish Movva, Founder & CEO of CarePredict

 

Technology will help ease, but not replace, rising workforce demand in long-term care: UCSF study

A just-published research paper by researchers at the University of California, San Francisco Health Workforce Research Center on Long-Term Care, has come to the not entirely unsurprising conclusion that the current technology targeted to the LTC area is helpful but won’t displace any workers from their jobs in the immediate future. The qualitative study evaluated 13 current health tech technologies in 14 areas for their potential impact on the care of older persons as it affects LTC workforce recruitment, training, and retention. 

Some key findings were: 

  • Technology will not even come close to replacing the LTC workforce. At most it will aid LTC workers.
  • Tools such as data collection and remote patient monitoring systems that distribute data to the care team can improve staff’s understanding of client behavior and manage day-to-day tasks
  • Technology can also address workforce recruitment, retention, and staffing efficiency, such as predictive analytics used in identifying candidate suitability, improved staff management in shift scheduling, work location, and clientele, and real time location tracking, can improve the work environment
  • Technologies that monitor health and activity measurements, integrating with predictive modeling, can benefit clients, family caregivers, and care teams, but may suffer from complexity and duplication in their category. 
  • Educational tools also improve care delivery by instructing on proper caregiving techniques, increasing knowledge on medical or behavioral conditions, and by promoting sympathy/empathy

Some of the barriers included:

  • It comes at a cost which LTC is reluctant to pay
    • Initial and ongoing cost with lack of third-party Medicare/private reimbursement
    • Dependence on unattractive long term subscription-based models 
  • Threats to privacy and the security of health data
  • Potential differences in product specificity or acceptance among diverse racial and ethnic groups
  • Technology lacking user-centered design and not developed/tested in conjunction with real-world LTC 
  • Funding: only two US VCs fund LTC tech is a bit of an exaggeration, but the pool of interest is shallow nonetheless

The overall conclusion struck this Editor as less than enthusiastic, perhaps because We’re Not There Yet and it’s still so far away.

The appendix lists the 13 companies surveyed with summaries of each health tech company interviewed: Alma’s House (Sweden), Arena (staffing/recruitment), Canary Health (education/caregiver education), CarePredict (wearables/alert monitoring), Clear Care (management). Embodied Labs (education), Intuition Robotics (ElliQ), GrandCare (monitoring/client engagement), Honor (staffing), La
Valeriane (documentation), LifePod (voicetech/monitoring), UnaliWear (wearables/monitoring), VisibleHand (documentation/EHR).

The study was supported by the Health Resources and Services Administration (HRSA) of the US Department of Health and Human Services (HHS).com. UCSF summaryThe Impact of Emerging Technologies on Long-Term Care & the Health Workforce (full text)  Hat tip to Laura Mitchell of GrandCare via Twitter

Fall prevention: the technology–and Dutch–cures

The ‘Holy Grail’ of fall detection is, of course, fall prevention. The CDC statistics for the US are well known: One in four Americans aged 65+ falls each year. Every 19 minutes, an older adult dies from a fall. Falls are the leading cause of fatal injury and the most common cause of nonfatal trauma-related hospital admissions among older adults–2.8 million injuries treated in emergency departments annually, including over 800,000 hospitalizations and more than 27,000 deaths. In 2014, the total cost of fall injuries was $31 billion. In the UK, AgeUK‘s stats are that falls represent the most frequent and serious type of accident in people aged 65 and over, the main cause of disability and the leading cause of death from injury among those aged 75+. 

The technology ‘cures’ as noted in this NextAvenue/Forbes article centers around predicting if and when a person will fall.

  • The ‘overall’ approach, which is constant monitoring of ADLs through activity sensing and modeling/machine learning to detect early signs of decline or health change. Companies in this area are Care Innovations’ QuietCare (sensor arrays) and CarePredict (wrist worn).
  • Gait detection. Relatively small changes in gait and walking speed are an accurate, fast, and straightforward indicator of fall risk. Ten years of research performed at TigerPlace in Missouri showed that people whose gait slowed by 5 centimeters per second within a week had an 86% probability of falling during the next three weeks. Shortening of stride had a 50 percent probability of fall within three weeks.
  • Read the brain. Research at Albert Einstein School of Medicine in NYC indicates that in otherwise high-functioning older people, high levels of frontal brain activity while walking and talking can predict higher long term fall risk, up to 32 percent.
  • Balance impairment. Tests using VR to simulate falling in healthy subjects and tracking their muscular response also could be used to roadmap a person’s balance impairments and future fall risk–along with training and targeted physical rehabilitation.

The Netherlands has taken this last point and gone ‘low tech’ with physical training courses that teach older adults both not to fall and to fall correctly if they do. Students negotiate obstacle courses and uneven surfaces, then learn to fall properly on thick inflated mats. Many of those attending use walkers or canes, but complete the courses which reduce the fear of falling or getting up–and provide both fun and socialization. The courses have become popular enough that they are government rated with insurance often defraying the cost. New York Times

Some quick, cheerful updates from Welbeing, CarePredict, Tunstall, Tynetec, Hasbro, Fitbit

It’s Friday, and in search of cheerful topics, here are some updates on doings from telecare, telehealth, and related companies we’ve recently noted on TTA:

Welbeing‘s opened a new head office at Technology Business Park in Moy Avenue in Eastbourne….CarePredict‘s AI for ADL system using the Tempo wearable has new implementations at LifeWell Senior Living’s community in Santa Fe, New Mexico (their third with CarePredict) and a three-year commitment with the Avanti Towne Lake community, Cypress, Texas. Dave Muoio has an interview with CEO Satish Movva on Mobihealthnews….Tunstall is partnering with Milpitas, California-based noHold’s Albert bot to create a virtual assistant for Tunstall’s mobile Smart Hub product, currently in Australia and in trials in Europe and the USA….Tynetec (advert above) has been closely associated and fundraised with the Dementia Dog Project and DogsforGood. An article in the Express highlights both in the beneficial role of pets with Alzheimers and dementia sufferers…. In robotic pet news, Hasbro is upgrading its ‘Joy for All’ companion pets through a Brown University research program, Affordable Robotic Intelligence for Elderly Support (ARIES) to add medication reminders, basic artificial intelligence, and more (Mobihealthnews)….Fitbit continues its march to a clinicalized product touting diabetes management partnerships with Medtronic and DexCom, plus clinical trials detecting sleep apnea through its SpO2 sensor. 3rd quarter sales were up 23 percent to $244 million and 40 percent from repeat purchasers, but they took an $8 million loss from a distributor (MedCityNews).

Themes and trends at Aging2.0 OPTIMIZE 2017

Aging2.0 OPTIMIZE, in San Francisco on Tuesday and Wednesday 14-15 November, annually attracts the top thinkers and doers in innovation and aging services. It brings together academia, designers, developers, investors, and senior care executives from all over the world to rethink the aging experience in both immediately practical and long-term visionary ways.

Looking at OPTIMIZE’s agenda, there are major themes that are on point for major industry trends.

Reinventing aging with an AI twist

What will aging be like during the next decades of the 21st Century? What must be done to support quality of life, active lives, and more independence? From nursing homes with more home-like environments (Green House Project) to Bill Thomas’ latest project–‘tiny houses’ that support independent living (Minkas)—there are many developments which will affect the perception and reality of aging.

Designers like Yves Béhar of fuseproject are rethinking home design as a continuum that supports all ages and abilities in what they want and need. Beyond physical design, these new homes are powered by artificial intelligence (AI) and machine learning technology that support wellness, engagement, and safety. Advances that are already here include voice-activated devices such as Amazon Alexa, virtual reality (VR), and IoT-enabled remote care (telehealth and telecare).

For attendees at Aging2.0, there will be substantial discussion on AI’s impact and implications, highlighted at Tuesday afternoon’s general session ‘AI-ging Into the Future’ and in Wednesday’s AI/IoT-related breakouts. AI is powering breakthroughs in social robotics and predictive health, the latter using sensor-based ADL and vital signs information for wellness, fall prevention, and dementia care. Some companies part of this conversation are CarePredict, EarlySense, SafelyYou, and Intuition Robotics.

Thriving, not surviving

Thriving in later age, not simply ‘aging in place’ or compensating for the loss of ability, must engage the community, the individual, and providers. There’s new interest in addressing interrelated social factors such as isolation, life purpose, food, healthcare quality, safety, and transportation. Business models and connected living technologies can combine to redesign post-acute care for better recovery, to prevent unnecessary readmissions, and provide more proactive care for chronic diseases as well as support wellness.

In this area, OPTIMIZE has many sessions on cities and localities reorganizing to support older adults in social determinants of health, transportation innovations, and wearables for passive communications between the older person and caregivers/providers. Some organizations and companies contributing to the conversation are grandPad, Village to Village Network, Lyft, and Milken Institute.

Technology and best practices positively affect the bottom line

How can senior housing and communities put innovation into action today? How can developers make it easier for them to adopt innovation? Innovations that ‘activate’ staff and caregivers create a multiplier for a positive effect on care. Successful rollouts create a positive impact on both the operations and financial health of senior living communities.

(more…)

Fall risk in older adults may be higher during warm weather–indoors

A new study contradicts the accepted wisdom of ‘when’ and ‘where’. Fall risk for older adults peaks in the winter, with outdoor falls in the ice and snow. Wrong. A new study presented at the recent Anesthesiology 2017 meeting of the American Society of Anesthesiologists found that hip fractures peaked during the warmer months at 55 percent.

  • The leading months were May (10.5 percent), September (10.3 percent), and October (9.7 percent)
  • Over 76 percent of those fractures occurred indoors while tripping over an obstacle like throw rugs or falling out of bed
  • Outdoor fractures in warm months were led by trips over obstacles, with the second and third leading causes being struck by or falling from a vehicle (!) or falling on or down stairs

The study sampled 544 patients treated at The Hospital of Central Connecticut for hip fracture from 2013 to 2016, with warm months defined as May 1 through October 31. Study author Jason Guercio, MD, MBA concluded that “Given the results of this study, it appears that efforts to decrease fall risk among the elderly living in cold climates should not be preferentially aimed at preventing outdoor fractures in winter, but should focus on conditions present throughout the year, and most importantly on mitigating indoor risk.” For caregivers, another reason why hazards in walking areas have to be reviewed and minimized.

The information provided does not give any indication as to the patient activity when the accident happened. There was also no correlation with health conditions or time. For instance, other studies have pointed out that a person rising out of bed in the morning has a change of blood pressure (high and low), and in the middle of the night, that person may be half-asleep. 

Where does technology come in? Getting ahead of the curve via gait analytics to alert for changes in gait and difficulty in walking. Noticing those changes could lead to proactive care and prevention. But as of now, those systems are either in test (Xsens MVN BIOMECH, WiGait TTA 4 May, Carnegie-Mellon TTA 23 May 16, Tiger Place MU TTA 29 Aug 15) or in early days in assisted living (CarePredict)–which doesn’t much help older adults at home. ASA release, McKnight’s Senior Living

A New Year’s Resolution, ADLs and a new care option

Here are three items that are each important and have hit my screen in the past couple of days – sadly, try as I may, I’m struggling with a common linking theme.

The first, that the 3G Doctor alerted me to, is a simply brilliant talk by Telcare‘s CEO Dr Jonathan Javitt at the Technion Social-Mobile-Cloud Meets Medicine Conference on the 17th December 2013. We’ve all made the arguments that technology enables the genuinely continuing care that long term conditions require, rather than the episodic care our health service is set up to provide, and that technology ensures that patients have clinical support 24/7 rather than in the brief period the doctor or nurse sees them.  However Dr Javitt brings all the arguments together to make such a powerful case that the only sensible way to treat long term conditions is to use technology to help the patient that anyone opposing it might as well try to argue that the earth is flat. As a result I have decided that my New Year’s resolution this year will be no longer to rise to the challenges of the naysayers. (I wonder how long I can keep it.)

The second item is a new take on monitoring activities of daily living (ADLs). For those new into telecare, continuous ADL monitoring looks a brilliant way of picking up an early decline in cognitive or physical decline, often well before symptoms show up in a change of vital signs or response to questions. The challenge though is whether the computer analysing the ADLs is smart enough to cope with activities such as the invasion of the grandchildren, or can cope with multiple occupancy. So it’ll be interesting to see how well CarePredict’s service is received. This uses a bracelet to track someone being cared for, rather than relying on PIRs or similar sensors as many other ADL systems do. Of course, like falls detectors, the problem with wearables is that people take them off, although the mHealth News item claims that ‘seniors’ like the bracelets.

The third item is a BBC item on the attractions of care homes in countries where the cost of living is lower, such as Thailand, which does feel a tad mercenary, although where there is genuine reverence for older people the quality of care can be excellent, and recent revelations suggest that care for older people in the UK is hardly without its problems. A combination of Skype and cheap flights certainly means that it is possible to keep in touch regularly. If it gets to be considered a viable option, it will certainly complicate the economics of technology to stay at home vs care home.

Hat tip to Prof Mike Short for alerting me to the BBC item.