Exploring machine learning in the cloud with AWS

There has been a resurgence of interest in machine learning in recent years. The ability to uncover hidden patterns and derive insights from vast, complex datasets is exceptionally valuable in an age where sensors are omnipresent and storage is cheap. Machine learning’s current status as the technology ‘buzzword of the year’ does not diminish the enormous potential of this technology in applications ranging from healthcare to self-driving cars.

Despite this, there are still considerable complexities in the design and implementation of a machine learning solution. Computational power is one of the core challenges; training a machine learning algorithm involves massive numbers of floating point operations, which regular computers perform relatively slowly and inefficiently. Cloud-based machine learning solutions can address this by offering access to highly optimised hardware that can perform the calculations at a fraction of the time and energy required by a traditional computer. AWS P3 instances are a great example of these specialised machines, providing up to one petaflop of floating point performance – meaning the machine can complete 1,000,000,000,000,000 floating point operations per second. This is around 1000 time faster than the average consumer PC. Cloud solutions can also reduce the cost and complexity of implementing a machine learning solution by providing developers with a pre-provisioned and configured platform to work with.

AWS recently hosted a seminar in Melbourne providing an overview of their machine learning platforms and capabilities, which I attended to explore opportunities for integration with Umps Health’s analytics platform. The session provided an excellent overview of the AWS offerings, which are broadly divided into three categories:

  1. Application Services
    These are API-based services for specific language and visual applications. These services are a simple and relatively low cost way to perform common machine learning tasks like image recognition, natural language processing, and translation. The pricing is based on the number of API requests made.
  1. Platform Services
    The most important service in this category is SageMaker, a managed service which enables a machine learning solution to be built, trained, and deployed using a simple modular interface. Sagemaker eliminates the complexities involved in platform implementation and enables users to focus on more important aspects – the data, algorithms, and insights. SageMaker’s pricing structure is based on usage time, billed per second.
  1. Frameworks and AMIs
    AWS also offers specialised machine learning Amazon Machine Images (AMIs). These are virtual machines running on optimised hardware, which are preconfigured with a range of popular machine learning tools like Apache MXNet, TensorFlow, and PyTorch. This option gives developers the maximum amount of control over the machine, but requires a greater level of technical knowledge to get the best results. The pricing model is based on hours of virtual machine uptime.

It’s important to note that (as of March 2018), SageMaker is only available in the US and EU regions. This means that users outside these regions (in Australia, for example) can still use SageMaker, but their data must be transferred to one of these regions for processing – a major problem for personal or sensitive data, which is generally subject to strict privacy laws and limitations on cross-border disclosures. Machine learning application services and AMI instances are already available worldwide, so most users will be able to run analytics with these services in the region in which their data was collected.

The seminar concluded with a demonstration of AWS DeepLens, a camera that analyses video using deep learning in real time and securely integrates with other AWS services. The accuracy of its object recognition algorithms was generally very good, and as a proof-of-concept and development tool DeepLens is impressive.

By eliminating the need for prohibitively expensive hardware and streamlining the implementation process, cloud-based machine solutions can significantly lower the cost and complexity of creating a powerful machine learning platform. However, as with any solution that handles personal data, privacy and security must remain a top priority.

February 2018 product update: An updated dashboard and interactive alerts system

This article was co-authored by Vincent Tseng, a Principal Engineer at Umps Health, and Adam Jahnke.

As many businesses went on break over summer, we were busy implementing a number of new features developed based on your feedback. As always, if there’s anything else you would like to see please contact us on info@umpshealth.com. We’d love to hear from you.

Umps Health’s dashboard now displays when appliances like the TV and bedside lamp are turned off

We collect a lot of data. The challenge has always been how do we make sense of that data, and present insights to family and older people that allow them to take action.

To simplify our dashboard, up until now we’ve only displayed when appliances were activated but not when they turn off. This isn’t a problem for the kettle, microwave or refrigerator as these appliances are only on for a few minutes, so the ‘on’ and ‘off’ action are associated with the same activity. However, it is useful to know when a person has switched off the TV or bedside lamp (or if they haven’t!), as these sometimes occur hours after switching them on.

Because of this, we’ve recently upgraded our dashboard to display ‘off’ events for certain appliances, providing a a clearer overview into an older person’s routine. Users will now notice these events appearing as hollow circles in the Activity Summary screen. Our alerts system (which has always considered the off-states) will continue to account for changes in these activities.

We’ve introduced an interactive alert system for people to manage their own health

Carers Australia define carers as “people who provide unpaid care and support to family members and friends who have a disability, mental illness, chronic condition, terminal illness, an alcohol or other drug issue or who are frail aged”. Collectively, they undertake more than 1.9 billion hours of care in Australia every year. If we were to replace that care, it would cost 60.3 billion dollars, making carers the backbone of Australia’s society.

From our work with carers, we decided that our system should focus on supporting the relationship between an older person and their family member rather than replacing it. We design our technology to empowering family members to deliver care more effectively and balance caregiving with living.

However, over the past 6 months we’ve received an increasing number of enquiries from people managing their own care. These people are highly independent, and don’t want alerts going to their family members before they’ve had a chance to manage their own care.

To accommodate for this, we’ve introduced a new alerts system. In this system, people managing their own care can interact with our alerts system by text message before an alert is escalated to a family member or friend. For example, if a person hasn’t boiled the kettle that morning they would receive the message to their phone. If they are OK, they can reply to text message and the alert will be cancelled. If there’s no response to the message with a defined period (we typically set this to 15 minutes), alerts will be sent to a nominated family member.

The post February 2018 product update: An updated dashboard and interactive alerts system appeared first on Umps Health.

Applying design principles to end-of-life care: Could you design One Good Death?

Meet Norman.

Norman is 89 years old, and has lived in the same house for 50 years. Norman’s daughter visits him regularly with her dogs, but after his wife passed away last year he’s become more and more socially isolated. Now, his day revolves around medical appointments. Norman has a predicted lifespan of 6 months.

How would you design a good death for Norman?

There’s a workshop for that.

Recently I was invited to join a workshop to apply the principles of design to end of life care and ‘design’ one good death. The One Good Death Workshop was hosted by Bolton Clarke and RMIT as part of the Engaging for Impact program, which this year has a big focus on the future of care.

Matiu Bush, Innovator in Residence at Bolton Clarke, opened the session by explaining Bolton Clarke’s approach to mapping people’s end of life journey, and outlining the goals of the workshop.

“The concept of a good death is one that recognises the whole-of-life contribution of people living in aged care, or in the community, as they near the end of life. Through focusing on one good death we hope to break this wicked problem into a number of smaller, more personal experiences that participants can address through empathetic design approaches. By designing one good death we hope to promote an approach that can be scaled to help many families” – Matiu Bush

As my fellow participants were introduced I began to notice the diversity among people in the room. There were artists, architects, designers, academics, aged care providers, funeral directors, startups, advocates and representatives from peak bodies and the Victorian Government. I suddenly had an appreciation of how people’s end of life journey involves navigating a complex system.

Adam Jahnke at the One Good Death workshop

Adam Jahnke at the One Good Death workshop hosted by RMIT and Bolton Clarke

Using empathy to design One Good Death

To design one good death, the group used a process developed by Leah Heiss, a Melbourne based designer and lecturer at RMIT. We used different tactile shapes to represent Norman’s goals, roadblocks to achieving those goals and potential workarounds. By the end of the workshop we had developed a ‘pathway’ to a good death for Norman.

The greatest workarounds were achieved when teams stopped looking at end of life through a “medicalised” paradigm and instead empathised with the personas and their needs. For some, a good death involved appointing a professional ‘death champion’ to advocate for them. For others, it required a facilitated family conference alleviating the worry around family conflict. For one of the personas, it was simply being assured his cat would be looked after when he died. Technology was referenced a few times, but it was always with the intent of enabling a specific outcome for a person.

How would your loved ones like to die?

Using personas and design principals is a different approach to developing end-of-life care, which was a highly debated topic in 2017. Last year, the Victorian Government passed legislation to enable assisted dying for a small group of Victorians suffering from an incurable illness and living in intolerable pain. Among my peers, I’ve noticed that many people come to support the legislation after witnessing the death of a loved one. In fact, it was the death of Premier Daniel Andrews’ father that changed his mind on assisted dying. It’s confronting, but considering how the people we love would like to die fosters empathy with those in end-of-life care. As the One Good Death workshop showed, this might be the key to designing and ‘scaling’ one good death.

Congratulations again to RMIT and Bolton Clarke. The workshop was truly an innovative approach to end of life care; I’m looking forward to seeing the outcomes and bringing the process of ‘roadmapping’ the ageing experience to Umps Health’s product team.

The post Applying design principles to end-of-life care: Could you design One Good Death? appeared first on Umps Health.

How to make sure your thermostat doesn’t attack your toaster: Security in the Internet of Things

After a slow start, Internet of Things (IoT) solutions for the home are experiencing a surge in popularity. Recent research by Telsyte found that more than 40% of Australian households currently have at least one IoT device, with more than 150 million devices expected to be in operation by 2021. As smart home devices become more widespread, however, security becomes an increasing concern; 2016 research by Deloitte found that 80% of consumers did not feel well-informed about security risks and 13% cited it as the main reason for not adopting IoT solutions at home.

Data and device security is vitally important in any IoT solution, and the responsibility for this must lie with the companies developing these solutions. Facing an ever-increasing range of threats, what can a company do to ensure its users’ data and devices are secure? While cybersecurity is a complex topic, the most important aspects are quite straightforward:

1. Design for security in every layer

In a typical smart home solution data is constantly flowing between connected devices, smart hubs, cloud platforms, mobile devices and web apps. Finding the best way to move data around is one of the core challenges of IoT; while cost, interoperability and energy efficiency constraints often dictate which technologies are used, it is critical to consider security here too.

Controlling wireless network access in the home, exclusively using encrypted communication protocols, and restricting access to cloud platforms will go a long way towards ensuring the safety and integrity of the solution. Secure data storage is also essential – don’t make the same mistakes as CloudPets, a line of internet-connected teddy bears which in 2017 was found to be storing customers’ personal information and voice recordings in publicly accessible Amazon cloud storage.

2. Plan for the future

Sooner or later, vulnerabilities will be found in any system, often in areas beyond your control. In October 2017, for example, Belgian researchers revealed the KRACK attack against WPA2, a ubiquitous security protocol used in all modern WiFi networks. This attack meant that every secure WiFi network became vulnerable due to a previously unknown flaw in WPA2. It is rarely possible to predict these incidents, but once a vulnerability is exposed a solution will generally be published very quickly. It is therefore vital that every aspect of an IoT solution can be rapidly upgraded in response to new threats.

3. And finally – get the basics right

In October 2016 the Mirai botnet launched multiple massive denial-of-service attacks, preventing users from accessing a range of major websites including Amazon, Airbnb, Twitter and Spotify. This incident was particularly significant as it was the first high-profile attack carried out by millions of infected IoT devices such as cameras, printers and thermostats. Mirai was able to infect all of these devices by simply attempting to log into them with common default usernames and passwords. In this case, the most basic level of security – setting the username and password of the devices to something other than ‘admin’ – would have been enough to keep these devices safe.

In addition to changing default credentials, every organisation working with IoT solutions should also be:

  • Ensuring the entire software stack is up to date at all times
  • Protecting your cloud credentials
  • Implementing a strong password policy, and never storing or transmitting users’ passwords in plaintext
  • Closely monitoring who has access to your data at all times. For example, is your sensor data being sent to someone else’s cloud too?
  • Raising employee awareness about the importance of security

These best practices are not new or revolutionary, but without them the strongest security architecture and most advanced cryptography in the world won’t keep a system secure.

IoT solutions for the home inherently deal with huge amounts of personal data; protecting this data and ensuring device integrity must be a priority. As the number of connected devices increases and threats multiply, it is critical that the developers and providers of IoT solutions be both transparent and proactive in their approach to cybersecurity.

The post How to make sure your thermostat doesn’t attack your toaster: Security in the Internet of Things appeared first on Umps Health.

Workplace gender equality: The problems to fix are in our heads

Sarah Goss, Head of Innovation at Ericsson Australia & New Zealand, Founding Director of Umps Health.
This article is reposted from Sarah’s website.

In April this year, Chief Economist of The Australia Institute, Dr. Richard Denniss, gave an address to the Victorian Women’s Trust on ‘The 3 Big Lies Holding Women Back’.  I have watched it several times, and shared it many more.  One of the uncomfortable truths he spoke aloud included this:

The inequality in the Australian labour market is not some accident. It’s not some undiscovered problem that is yet to percolate to the top of the political agenda because of the lack of evidence. A lot of powerful people in Australia are entirely happy with it.  That, is what you’re up against.”

And Australia is not alone.  In no country in the world today are women equal.  In fact, in the 2017 Global Gender Gap Index Report published last month, the World Economic Forum projects it will take 217 years to reach gender equality in our workplaces. That means it will be another 7 generations before we see workplace gender equality – or, until my great, great, great, great, great, great grandchildren enter the workforce.

I would probably be forgiven for throwing my hands up in the air in despair, but I prefer to adopt a systems thinking approach; it permits me to be a little more optimistic.  Systems change takes time – sometimes generations – but every action we take today can create positive knock-on effects, moving towards the dismantling of interrelated structures that maintain the status quo in our workplaces.

There are many barriers to tackle, but I’ve found a few prickly ones.  It’s because they are the ones that most often make people bristle that I know I am hitting on truths we can’t ignore.

We all need to face up to challenging our own thinking and preconceptions, and commit to 3 fundamental changes in our collective mindset.  The problems to fix start in our heads.

Mindset shift #1: Give up the belief that your organisation is a meritocracy

As the numbers crunched by industry expert Conrad Liveris show, women comprise only 4.5% of all ASX 200 CEOs.  There are more CEOs of ASX 200 companies called either John, Peter or David than there are women CEOs in total.  Liveris’ analysis further reveals that Australian leadership is blindingly white.  Only 3.9% of ASX executives have non-European backgrounds.

The Australian Human Rights Commission also conducted research into the cultural diversity of Australia’s leadership.  It found that 95% or thereabouts of senior leaders in Australian business, politics, government, and civil society are white.

As Liveris points out, “straight, white, able-bodied men aged 40-69 years, which represents the majority of Australian leadership, are 8.4 per cent of the population.”

If we are to believe that our companies, institutions and other organisations see the valuing, reward and elevation of people based on their merit, then we are to believe that straight, white, middle-aged men have a monopoly on merit.  We know this isn’t true.

Now, I am not saying that men in these roles haven’t worked hard to get to where they are, or that they are undeserving of their positions.  I am not attempting to detract from their competence, work ethic, commitment or otherwise.  But, there are clearly other factors at play.

It is time to drop the myth that our workplaces are meritocratic.  No organisation is a meritocracy until it has accounted for all of its inherent and unspoken biases.

Mindset shift #2: We will not have gender equality in our workplaces, until we have gender equality in our homes

If the stats overwhelmingly show us that men are the leaders at work, then in comparable magnitude they also prove undeniably that women are the leaders at home.

Research such as the Melbourne Institute’s HILDA survey demonstrates that there are still very entrenched conventions about the differing roles between the genders.  The study shows that even many women subscribe to the construct of ‘male breadwinner, female homemaker’, reflecting just how strong the societal conditioning and accepted norms are around women’s place in Australia.

According to the federal government’s Workplace Gender Equality Agency (WGEA), three quarters of all unpaid work is done by women.  This includes activities such as cooking, cleaning, washing, gardening, home maintenance, looking after children, caring for the elderly or a family member with a long-term health condition or disability, and doing voluntary community work.

The WGEA quantifies the difference between men and women’s time spent on these activities to an average ‘gender time gap in unpaid work’ of 2 hours and 19 minutes per day.  Or, for every one hour of unpaid work that men do, women do almost double.  Professor Mark Wooden, Director of the HILDA survey says,

“Women still take on the bulk of household chores and assume greater child care responsibilities. This means that men are spending more time at work building experience, skills and networks.  Until we see a greater sharing of roles and responsibilities at home, women, on average, will always be disadvantaged in the labour market relative to men.”

We are far from having equality in our homes.  If we do not acknowledge that we must address women’s unpaid workload, we will not see more equal participation and representation of women in paid work.

Mindset shift #3: You cannot claim free will and blame women’s choices

At this point, some may be tempted to rationalise that it’s women’s choice to stay at home with the kids, or to cut back to a part-time role, or to take on more basic work, or to be naturally drawn to lower-paying professions, or to not put themselves forward for promotions because they are not that career-oriented, or… the list goes on.

However, when the gender stereotype of the role of women is so engrained in our culture, we have to recognise the impact of context.

This week I had the privilege of meeting Professor Michelle Ryan from the University of Exeter, UK, and hearing her present at the Future of Work Conference on her research into how context constrains women’s work choices.  Immediately following, I joined her and other industry leaders on a panel discussing ‘Beyond Token Diversity’.

Ryan has found a direct correlation between women’s feelings of fitting in and belonging in an organisation, and how ambitious and committed they are at work.  Women must see potential benefit and reward for their ambition, motivation, risk-taking, and sacrifices at work.  Their ‘choices’ do not occur in a vacuum, they are shaped and constrained by organisational and social context.

In this way, Ryan’s evidence proves it is too simplistic to accept the popular belief that women opt for a work-life balance based predominantly on the practicalities of time and how to juggle it between work and family.  Just as important as time is how a woman feels about the workplace.  She explains,

“It’s not just about time – how many hours they work – it’s feeling like you’re similar to others before you; if people like you have made it.  Being similar to successful others reinforces your identity – who you have to be to be successful at work, is compatible with who you are at home.”

Time + identity = ‘work-life compatibility’.  To maximise compatibility of a workplace for women, organisations need to think about how they can foster a sense of belonging so that women can see the possibility to thrive as their whole selves at work.

From behind the hollow defence lines

If we want to see genuine change in workplace gender equality, we need to question our assumptions and notions of ‘merit’ and ‘women’s choice’, and address the inequality in the division of unpaid work in our homes.  To stand behind false beliefs to the contrary will only continue to hinder our progress against inequality.

And 217 years is too long for any of us to wait.

The post Workplace gender equality: The problems to fix are in our heads appeared first on Umps Health.

Dreamstarter Campaign Launch

Today is the day – our crowdfunding campaign has just gone live! In partnership with Carers Australia and as part of the ING Dreamstarter Program, for the next 30 days we’re asking for your help to support people caring for loved ones return to school and work.

2.8 million Australians are caring for loved ones, and many are spending 20+ hours delivering care. We know that Umps Health can empower these people to continue living their lives while delivering high quality care by allowing them to always be digitally present with their loved ones.

The funds we raise will fund our hardware costs for direct deployments into these people’s homes, allowing us to take an idea that worked for my family to the world. We’ve some great rewards on offer that will be delivered just in time for Christmas, as well as the opportunity to sign up to our service on early release.

In an incredible show of support, ING will match every dollar we raise for the first $14,500 – so please show your support and become a Dreamstarter with us here!

The post Dreamstarter Campaign Launch appeared first on Umps Health.

Caring for ageing parents: flexible work and technology innovation

Sarah Goss, Head of Innovation at Ericsson Australia & New Zealand, Founding Director of Umps Health
This article is reposted from Sarah’s website.

“I don’t know how you do it!”, is a remark I often hear when people learn that I am an executive at Ericsson, Founding Director of a start-up, and a mother and primary carer for two young kids.

But balancing work and family life goes beyond caring for children.  With our ageing population, the demand for people to provide care for a family member who is in old age or has a serious illness or disability has risen sharply.

Care to work differently

According to the Australian Bureau of Statistics (ABS), 2.7 million Australians are ‘informal carers’.  That’s one in eight Australians delivering an average of 13 hours of care per week to an older family member, or a family member in need due to disability or illness.  Put another way, Australia’s informal carers are working full-time 2 days a week delivering care.  Unpaid.  Meaning their work is not formally recognised for the $60.3 billion of economic value it contributes to our national economy.  Like me, even if you’re not an informal carer yourself, you know someone who is.  They are the backbone of our healthcare system and community.

The burden of providing these 1.9 billion hours of informal care every year is disproportionately shouldered by women.  More than two thirds of informal carers are women, with most aged between 55 and 64.  I use the word ‘burden’ because 44% of working-age primary carers have opted out of the workforce (ABS), and the prevalence of depression is 67% higher among these high-intensity carers than non-carers.

This over-reliance on informal carers and the personal toll it takes on them is in urgent need of redress.  There is no silver bullet, but employers can play a role to improve support for ‘working carers’.

More job-share, yeah yeah!

It can be very difficult for informal carers to get back into the workforce if they’ve opted out as nearly half of them do.  Working carers often seek part-time work to better enable them to balance work and care, but part-time jobs are hard to come by.  Informal carers are typically older workers, so without workplace flexibilities and quality part-time work options for especially older employees, balancing paid work with caring for family members such as spouses and ageing parents can be extremely challenging.

The struggle to cope with managing work and care is not dissimilar to the experience I have had working whilst also caring for my young children.  I have been working 3 days a week in senior roles for almost 7 years, since I returned to work after having my first son. I have worked in a straight part-time role, and in a job-share.  Job-sharing is where a full-time position is held by two people, and I believe there is more potential for this model of job design to work particularly well for working carers.

These are the benefits I can attest to from being a primary carer and a job-sharer:

  • You can switch off on your non-work days.  Your job-share buddy is on the job, so you are able to focus on your caring responsibilities.
  • You have built in redundancy.  You can shuffle days between you or cover for one another if emergencies or extra-ordinary events arise.  There is another ‘you’, so you’re not a single point of weakness. You can rest assured knowing that your customers, staff, colleagues and stakeholders will not be let down.
  • You can still work in a senior role despite working part-time. Most senior roles are full-time, so job-sharing opens up access to these roles even though you work part-time.  You can continue to progress in your career with minimal disruption, if any.
  • You can satisfy other drivers and interests that you have that caring maybe doesn’t fulfil. Being able to work, and in a flexible and manageable way, can contribute to your overall wellbeing by giving you an outlet for professional pursuits and relationships with workmates.  It can also help alleviate the pressure and loneliness that many carers face.

When looking at ABS data (Table 37.1), it can be concluded that most jobs held by working carers are suited to job-share.  Furthermore, there could be specific promise in the wider availability of job-share opportunities for the 39% of working carers who are Managers or Professionals – some 522,000 Australians – to support them in continuing their career at a senior level.

Some tech with that too, please

I have friends who are informal carers who worry for their older family member, or feel guilty that they are not there ‘enough’.  More flexible work arrangements can certainly help working carers, but in combination with new advances in smart technology, carers can be reassured of the welfare of their loved one even when they’re not physically present with them.

Umps Health provides one such smart technology solution.  It offers a wellbeing detection system so that as older people living at home alone use their everyday appliances, usage can be tracked to detect any abnormalities in routine and then raise alerts to carers.

And for informal carers searching for ‘career worthy’ part-time, job-share or other types of flexible roles, you can be matched with employers in the connected job market through online marketplaces such as Beam Australia and OneJob TwoMinds.

A better way

Flexible work arrangements such as part-time hours and job-sharing can encourage better physical, mental and financial health for informal carers.  Smart technology can empower them with information about their older loved one and provide greater peace of mind.  Given our population is ageing and the demand on informal carers is at breaking point, it is clear Australia needs to look at these and other ideas for a better way of care.

If you are a working carer and would like to share your story, I would love to hear from you.

Sarah Goss works in a part-time executive role as Head of Innovation at Ericsson Australia & New Zealand and is also Founding Director at HealthTech start-up, Umps Health.

Twitter: @SarahGossAU
LinkedIn: https://www.linkedin.com/in/sarahgossAU

The post Caring for ageing parents: flexible work and technology innovation appeared first on Umps Health.

How will artificial intelligence support care in the home?

The emergence of more powerful computers, near-ubiquitous mobile broadband and cloud computing has created exponential change across every industry and society. Increasingly, we are seeing the convergence of physical and cyber systems, what society has termed the Fourth Industrial Revolution, Industry 4.0 and the Internet of Things (IoT). The IoT, combined with advancements in machine learning and artificial intelligence will fundamentally change society in the next five years with the commercialisation of autonomous vehicles, sensor-based health monitoring and ‘smart-homes’.

The aged care services industry is neither exempt nor oblivious to this change. In February 2014, The Aged Care Industry Information Technology Council and Accenture jointly released a vision for information technology across aged care services. This vision identified five ‘ICT Pillars’ that will enable the industry to build capacity, and adjust to ‘the new environment’. In 2016, Flinders University was commissioned to build on this vision, and outline an IT Roadmap that would support the execution of the Aged Care Sector Roadmap. This work is currently underway, with extensive consultation occurring at the Information in Aged Care Conference in November 2016.

In this article, I outline how advancements in the IoT and artificial intelligence have the potential to support carers of older Australians, as well as the long-term sustainability of informal care. This was the topic of a recent presentation at the Melbourne Accelerator Program’s Social Impact Showcase, where Umps Health was were awarded the Gourlay-Trinity Impact Prize for ‘demonstrating exceptional knowledge in a social issue’.


To demonstrate the capability of IoT and artificial intelligence, it is helpful to think of the workflow of a carer in the following stages:

  1. Gathering data:In order to provide care, a carer must first obtain data about the wellbeing of an individual. This could be through phone calls or a visit to the person. Other examples are the Red Cross TeleCross service, which call more than 9000 Australians every day or the button people press once a day when they use a personal alarm service.
  2. Analysing and interpreting the data: This data must then be analysed in the context of past behaviour and health history. Does a missed phone call indicate someone has fallen in their home, or are they out for a coffee with friends? If a person says they are OK, are they actually OK or are they simply trying not to worry people? We might not think about this explicitly, but we are subconsciously making an assessment when there is an abnormality in the data.
  3. Determining a recommended course of action: If a carer interprets the data as unusual, they will evaluate options and decide on a course of action. This may be to make a visit to their relative, schedule an appointment with a doctor or to call emergency services.
  4. Implementing the course of action: After a carer has determined the appropriate course of action, the ‘caring’ begins. This is the intrinsically human component of caring, and could be as simple as physical or emotional comfort, or driving someone to the doctor. In the event they need to consult a third party, the entire workflow will commence once again.


It is estimated that the average household in Australia has more than 11 connected devices, and this is projected to grow to more than 29 devices before the end of the decade. Some examples of connected devices include refrigerators, TVs, cameras, lighting, fitness trackers and bathroom scales.

The proliferation of these ‘IoT enabled devices’ has generated huge amounts of data, which service providers can leverage to develop innovative consumer services. For example, knowing that an individual has used their kettle in the morning is evidence they have got out of bed, while understanding how someone uses their microwave provides powerful insights into their eating habits. In this way, the IoT has enriched our ability to collect data. 


Artificial intelligence is a domain in computer science that seeks to simulate human intelligence. A key principle of intelligence is the ability to learn, and while we haven’t yet created the sorts of artificial intelligence popularised in fiction (like Rosie from the Jetsons), you may be surprised at the amount of artificial intelligence you already interact with everyday. Netflix uses advanced algorithms to assess your interest in different movies and make tailored predictions on content you will enjoy. Pandora uses artificial intelligence to analyse complex characteristics in music and build custom playlists for us, and Siri is constantly learning how to more effectively respond to human language questions.

Even those 11 devices currently in Australian homes could generate hundreds of data points per day, making it extremely difficult for human intelligence (carers or aged care service providers) to generate insights. This is in contrast to artificial intelligence, which can analyse small abnormalities in the context of past behaviour of the individual, while crosschecking against anonymised data from hundreds of thousands of other people.

Imagine your father boils water at 2:43AM and again at 3:17AM. This is not entirely unusual for your dad, who often has a cup of tea after working late on his computer. However, a machine learning program assesses this in conjunction with the large number of lights that have been left on in the house, the fact that the computer hasn’t been used all evening and knowledge that your dad made toast half an hour earlier. Further analysis against population data suggests that this combination of anomalies could be associated with a loss of sense of time, an early symptom of dementia. It isn’t difficult to imagine a future where a program would then review your calendar against available appointments with your father’s GP, and make a tentative booking on your behalf.


As the capabilities of artificial intelligence grow, there is a social imperative to the adoption of artificial intelligence-based services. Firstly, artificial intelligence can help older people to age in place, making greater sense of large amounts of data to predict and potentially prevent an incident like a fall. Secondly, it can provide peace of mind to the 2.8 million Australians providing unpaid care each year, allowing them to participate in paid employment and reducing the physical and emotional stress associated with being an informal caregiver.


I am a believer in the power of artificial intelligence and the IoT because I use it to provide better care to my family. After designing a solution like the above for my grandpa, I founded Umps Health. Umps Health uses uses connected devices to collect data about how people interact with their appliances, like their kettle, TV and microwave. Our machine-learning platform generates tailored insights based on past patterns of behaviour, and raises an alert to carers when behaviour is out of the norm. With this technology, we can better inform carers when an incident (like a fall) may have occurred, or when someone’s behaviour demonstrates they are at heightened risk of an incident.

By reducing the time and effort required to collect data, interpret data and determine a recommended course of action, we hope to improve the lives of both those receiving care and those delivering it.

About the Author
Adam Jahnke is the founder and CEO of Umps Health, a company that uses machine learning to enable better living outcomes for older people. He has previously worked on smart city and smart home projects throughout Australia and South-East Asia. Adam is also studying a Master of Public Health at the University of Melbourne, with a research focus on international and intergenerational response to challenges associated with ageing.

For more information, contact Adam at adam@umpshealth.com.


The post How will artificial intelligence support care in the home? appeared first on Umps Health.