Digital Transformation
Showing results for 
Search instead for 
Did you mean: 

How can you use Artificial Intelligence in health management?


AI will impact every industry to some degree or other. But I think that it will eventually be used a great deal in health management (I say, “health management” and not “healthcare”, because I believe one of the key ways in which AI will help will be in helping us “keep out of hospital”; keeping us from falling ill).

Certainly if number of articles in The Times (of London) is anything to go by, healthcare will be the place that AI will be used. For example, just today, The Times carried an article entitled, “Artificial intelligence could create personalised prostate cancer therapy”.

Healthcare outside of hospitals

Personal health management

Before about fifty, we get the idea that we will die some day, but not now. Beyond fifty, we get that we will die, but the concept of quality of life as determined by our health starts to rear its head. This is when we start to monitor blood pressure, cholesterol, inflammation markers, fasting blood sugar and the like. I suspect that in years to come, we will look back on these health indicators as primitive as we get genome tests, biome tests (the bacterial makeup of our guts), and take a bimonthly full blood test at home. (This article, from the Economist, talks about the latest developments in self-blood testing) 


AI healthcare runners.png

How does AI play in all this? As with all IoT applications, collecting the data is the “easy first part” (this is a blog post I wrote on  the evolution to machine learning). It’s what we do with the data that really counts - the action we take from the data.

In the case of health, of course, this action might include an exercise regime, a change in diet, supplements or a personalized course of drugs. 

AI will be used to take our “digital health footprints” (genome, biome, blood tests, personal history, sleep patterns, exercise patterns, stress patterns) and infer “what is going on” inside us. 

And then, AI will be used to figure out, given our personal health footprint, age, personality and existing habits (e.g. we might be a runner, but with weak upper musculature), what is the best course of action. 

Now, some healthcare professionals find the concept of self-health monitoring to be a little dangerous. But the fact remains that the supplement industry is a multi-billion dollar market and growing, and many of the people who go to the gym, run, do yoga, pilates, meditation, take supplements and walk do so to keep themselves out of the healthcare system. 

AI-based triage from home

In the UK a system called Babylon has been in trial for a while now. It is an AI-based system that allows you to have a triage conversation with a healthcare professional. As of yesterday, this system will enjoy wider rollout across the UK.  

Babylon is not, of course, a replacement for a one-to-one doctor’s appointment, but is more a way of applying both AI and mobility to the triage part of a consultation, perhaps avoiding a physical visit to the doctor or hospital all together. 

Monitoring at home

Most people want to continue to live in their homes as long as possible rather than being hospitalized going into old persons’ care homes. 

But if a vulnerable person lives at home, they will need a high degree of monitoring and this can be very expensive. Throughout the world, there is a shortage of carers (caregivers). 

And so a number of efforts are under way to monitor people in their homes. One system uses an “edge detect” camera to record the motion of the person under observation. AI image analysis systems autonomously look for a “fall pattern” in the motion of the person’s edges. When this occurs, the alert the emergency services.

Another system uses the appliance-use pattern of the person under observation. The theory is that as we get older, our patterns become more strongly ingrained - “Fred always gets up around 7:30 and turns on the radio. He makes a cup of tea between 8:00 and 8:30”. AI-based systems can build up a model of Fred’s usual appliance-use patterns, and then monitor them for deviations. 

The first paragraph in this article is an example of a company offering such a monitoring system. The third example in that article talks about a platform for those with diabetes. 

Precision healthcare

Personalized treatment regimens

Machine learning allows us to “throw” data about the treatments given to patients and the outcomes, and from this, the machine learning system can start to learn what treatments work best for what sets of symptoms..

McKinsey estimates that machine learning  and tailored treatments could make a 5 to 9 percent saving. 

AI healthcare apple.jpeg

Personalized drugs

The same techniques can be taken one step further - the actual drugs used in the treatments can be personalized.  This is a wide-ranging article from the UK Guardian newspaper which includes a vision of 3D printed drugs, and micro-robots inside us dispensing drugs. It ends with a quote from professor Daniel Kraft from Singularity University, “soon we won’t wait for disease to happen. We’ll care for ourselves before we get sick.”  

There is a Chinese proverb, “no sickness, short life. One sickness, long life”; the sentiment being that if you have no illnesses you are cavalier with your health, whereas if you have an illness, you are careful with your health.  Maybe we will be able to update this saying to, “no bio marker, machine learning health prediction, short life. One bio marker, machine learning health prediction, long, high quality life”

McKinsey predicts that the US alone will save $2T to $10T through the use of tailored drugs.

Inside hospitals

Reducing hospital re-admissions

Hospital re-admissions are expensive. Machine learning is being used to predict those patients that are most at risk of re-admission. Machine learning is used because there are simply too many variables for humans to consider. And, with the advent of ever more detailed bio-monitoring (think FitBit on steroids), prediction becomes even more complex and out of the reach of humans. 

Once we know which patients are at risk of re-admission we can take proactive action and/or increase monitoring.

Registration and triage kiosks

Just like apps like Babylon are being used for self-triage at home, we can do the same thing once people arrive at the hospital. 

Augmenting healthcare professionals

There is a lot of talk in the press about how AI is “going to take your job”. While this may happen in some situations, I think it’s more likely that AI is going to augment and help you in your job, just like a tractor augments and helps a farmer. 

Help with caring for the elderly

In many countries the population is ageing and we are already suffering a shortage of carers (caregivers) for the elderly. For example it is estimated that by 2025, Japan will have a shortfall of 400,000 carers for the elderly. This video shows a robot being used for elderly care in Japan.

Help with diagnosis

A number of companies, including Google’s DeepMind group in London, are building systems that look at the digital footprint of patients and suggest the presence of potential medical issues. Think of these systems as ultra-diligent monitors that sit alongside the medical professionals looking for the footprints that suggest an issue might be present. 

AI healthcare graph.png

(This graph above shows how medical imagery will increasingly use AI. AI will “look” at the medical images, augmenting healthcare professionals. Look out for the first legal case in the US of an AI system being sued for misdiagnosis.)

Here is an example article that describes a system in the UK that helps doctors predict heart conditions. And here is another that describes the use of machine learning in Boston that predicts both heart problems and diabetes. 

Robot-assisted surgery

Robots are already being used to assist surgeons both where great precision is required and also for remote surgery where the surgeon is not remote.

Robots for cleaning, cooking, etc

While it’s early days, there is a lot of work being done on the use of robots for cleaning and cooking in hospitals and care homes. 

Related posts

Mike Shaw
Director Strategic Marketing
Hewlett Packard Enterprise

twitter.gif @mike_j_shaw
linkedin.gif Mike Shaw

Mike Shaw
Director Strategic Marketing

linkedin.gifMike Shaw

0 Kudos
About the Author


Mike has been with HPE for 30 years. Half of that time was in research and development, mainly as an architect. The other 15 years has been spent in product management, product marketing, and now, strategic marketing. .


AI being the latest in technology to improve the user experience across all sections is a sure head turner. Health Care is huge where knowing a persons previous health history or some cures is a big thing. We have already seen how big data has impaced health care industry.

Good informative post.



AI will be used to take our “digital health footprints” (genome, biome, blood tests, personal history, sleep patterns, exercise patterns, stress patterns) and infer “what is going on” inside us. 

And then, AI will be used to figure out, given our personal health footprint, age, personality and existing habits (e.g. we might be a runner, but with weak upper musculature), what is the best course of action. 


Do you really think AI will be able to understand personality?

Apr 24 - 25, 2018
Expert Days - 2018
Visit this forum and get the schedules for online HPE Expert Days where you can talk to HPE product experts, R&D and support team members and get answ...
Read more
June 19 - 21
Las Vegas, NV
HPE Discover 2018 Las Vegas
Visit this forum and learn about all things Discover 2018 in Las Vegas, Nevada, June 19 - 21, 2018.
Read more
View all