Next Five Years of Healthcare

This insight is related to all of you and your children and relatives. It is about the health and healthcare. I feel confident to envision the progress for five years, but cautious to guess for longer. Even next five years seem pretty exciting and revolutionary. Hope you will enjoy they pathway.

We have problems today

I will not bind this to any country, hence American readers will not find Obamacare, ACO or HIE here. I will go globally as I like to do.

The old industry of healthcare still sucks. It sucks everywhere in the world. The problem is in uncertainty of our [human] nature. It’s a paradox: the medicine is one of the oldest practices and sciences, but nowadays it is one of least mature. We still don’t know for sure why and how are bodies and souls operate. The reverse engineering should continue until we gain the complete knowledge.

I believe there were civilisations tens of thousands years ago… but let’s concentrate on ours. It took many years to start in-depth studying ourselves. Leonardo da Vinci did breakthrough into anatomy in early 1500s. The accuracy of his anatomical sketches are amazing. Why didn’t others draw at the same level of perfection? The first heart transplant was performed only in 1967 in Cape Town by Christiaan Barnard. Today we are still weak at brain surgeries, even the knowledge how brain works and what is it. Paul Allen significantly contributed to the mapping of the brain. The ambitious Human Genome project was performed only in early 2000s, with 92% of sampling at 99.99% accuracy. Today, there is no clear vision or understanding what majority of DNA is for. I personally do not believe into Junk DNA, and ENCODE project confirmed it might be related to the protein regulation. Hence there is still plenty of work to complete…

But even with the current medical knowledge the healthcare could be better. Very often the patient is admitted from the scratch as a new one. Almost always the patient is discharged without proper monitoring of the medication, nutrition, behaviour and lifestyle. There are no mechanisms, practices or regulations to make it possible. For sure there are some post-discharge recommendations, assignments to the aftercare professionals, but it is immature and very inaccurate in comparison to what it could be. There are glimpses of telemedicine, but it is still very immature.

And finally, the healthcare industry in comparison to other industries such as retail, media, leisure and tourism is far behind in terms of consumer orientation. Even automotive industry is more consumer oriented than healthcare today. Economically speaking, there must be transformation to the consumer centric model. It is the same winning pattern across the industries. It [consumerism] should emerge in healthcare too. Enough about the current problems, let’s switch to the positive things – technology available!

There could be Care Anywhere

We need care anywhere. Either it is underground in the diamond mine, or in the ocean on-board of Queen Mary 2, or in the medical center or at home, at secluded places, or in the car, bus, train or plane.

There is wireless network (from cell providers), there are wearable medical devices, there is a smartphone as a man-in-the-middle to connect with the back-end. It is obvious that diagnostics and prevention, especially for the chronical diseases and emergency cases (first aid, paramedics) could be improved.

I personally experienced two emergency landings, once by being on-board of the six hour flight, second time by driving for the colleague to another airport. The impact is significant. Imagine that 300+ people landed in Canada, then according to the Canadian law all luggage was unloaded, moved to X-ray, then loaded again; we all lost few hours because of somebody’s heart attack.

It could be prevented it the passenger had heart monitor, blood pressure monitor, other devices and they would trigger the alarm to take the pill or ask the crew for the pill in time. The best case is that all wearable devices are linked to the smartphone [it is often allowed to turn on Bluetooth or Wi-Fi in airplane mode]. Then the app would ring and display recommendations to the passenger.

4P aka Four P’s

The medicine should go Personal, Predictive, Preventive and Participatory. It will become so in five years.

Personal is already partially explained above. Besides consumerism, which is a social or economic aspect, there should be really biological personal aspect. We all are different by ~6 million genes. That biological difference does matter. It defines the carrier status for illnesses, it is related to risks of the illnesses, it is related to individual drug response and it uncovers other health-related traits [such as Lactose Intolerance or Alcohol Addiction].

Personal medicine is an equivalent to the Mobile Health. Because you are in motion and you are unique. The single sufficiently smart device you carry with you everywhere is a smartphone. Other wearable devices are still not connected [directly into the Internet of Things]. Hence you have to use them all with the smartphone in the middle.

The shift is from volume to value. From pay to procedures to pay for performance. The model becomes outcome based. The challenge is how to measure performance: good treatment vs. poor bedside, poor treatment vs. good bedside and so on.

Predictive is a pathway to the healthcare transformation. As healthcare experts say: “the providers are flying blind”. There is no good integration and interoperability between providers and even within a single provider. The only rationale way to “open the eyes” is analytics. Descriptive analytics to get a snapshot of what is going on, predictive analytics to foresee the near future and make right decisions, and prescriptive analytics to know even better the reasoning of the future things.

Why there is still no good interoperability? Why there is no wide HL7 adoption? How many years have gone since those initiatives and standards? My personal opinion is that the current [and former] interoperability efforts are the dead end. The rationale is simple: if it worth to be done, it would be already done. There might be something in the middle – the providers will implement interoperability within themselves, but not at the scale of the state or country or globally.

Two reasons for “dead interop”. First is business related. Why should I share my stuff with others? I spent on expensive labs or scans, I don’t want others to benefit from my investments into this patient treatment. Second is breakthrough in genomics and proteomics. Only 20 minutes needed to purify the DNA from the body liquids with Zymo Research DNA Kit. Genome in 15 minutes under $100 has been planned by Pacific Biosciences by this year. Intel invested 100 million dollars into Pacific Biosciences in 2008. Besides gene mechanisms, there are others, not related to DNA change. They are also useful for analysis, predicting and decision making per individual patient. [Read about epigenetics for more details]. There is a third reason – Artificial Intelligence. We already classify with AI, very soon will put much more responsibility onto AI.

Preventive is very interesting transformation, because it is blurring the boarders between treatment and behaviour/lifestyle/wellness and between drugs and nutrition. It is directly related to the chronic diseases and to post-discharge aftercare, even self aftercare. To prevent from readmission the patient should take proper medication, adjust her behaviour and lifestyle, consume special nutrition. E.g. diabetes patients should eat special sugar-free meal. There is a question where drug ends and where nutrition starts? What Coca Cola Diet is? First step towards the drugs?

Pharmacogenomics is on the rise to do proactive steps into the future, with known individual’s response to the drugs. It is both predictive and preventive. It will be normal that mass universal drugs will start to disappear, while narrowly targeted drugs will be designed. Personal drugs is a next step, when the patient is a foundation for almost exclusive treatment.

Participatory is interesting in the way that non-healthcare organisations become related to the healthcare. P&G produce sun screens, designed by skin type [at molecular level], for older people and for children. Nestle produces dietary food. And recall there are Johnson & Johnson, Unilever and even Coca Cola. I strongly recommend to investigate PWC Health practice for the insights and analysis.

Personal Starts from Wearable

The most important driver for the adoption of wearable medical devices is ageing population. The average age of the population increases, while the mobility of the population decreases. People need access to healthcare from everywhere, and at lower cost [for those who retired]. Chronic diseases are related to the ageing population too. Chronic diseases require constant control, interventions of physician in case of high or low measurements. Such control is possible via multiple medical devices. Many of them are smartphone-enabled, where corresponding application runs and “decides” what to tell to the user.

Glucose meter is much smaller now, here is a slick one from iBGStar. Heart rate monitors are available in plenty of choices. Fitness trackers and dietary apps are present as vast majority of [mobile health] apps in the stores. Wrist bands are becoming the element of lifestyle, especially with fashionably designed Jawbone Up. Triceps band BodyMedia is good for calories tracking. Add here wireless weight… I’ve described gadgets and principles in previous posts Wearable Technology and Wearable Technology, Part II. Here I’d like to distinguish Scanadu Scout, measuring vitals like temperature, heart rate, oxymetry [saturation of your hemoglobin], ECG, HRV, PWTT, UA [urine analysis] and mood/stress. Just put appropriate gadgets onto your body, gather data, analyse and apply predictive analytics to react or to prevent.

Personal is a Future of Medicine

If you think about all those personal gadgets and brick mobile phones as sub-niche within medicine, then you are deeply mistaken. Because the medicine itself will become personal as a whole. It is a five year transition from what we have to what should be [and will be]. Computer disappears, into the pocket and into the cloud. All pocket sized and wearable gadgets will miniaturise, while cloud farms of servers will grow and run much smarter AI.

Everybody of us will become a “thing” within the Internet of Things. IoT is not a Facebook [it’s too primitive], but it is quantified and connected you, to the intelligent health cloud, and sometimes to the physicians and other people [patients like you]. This will happen within next 5-10 years, I think rather sooner or later. The technology changes within few years. There were no tablets 3.5 years ago, now we have plenty of them and even new bendable prototypes. Today we experience first wearable breakthroughs, imagine how it will advance within next 3 years. Remember we are accelerating, the technology is accelerating. Much more to come and it will change out lives. I hope it will transform the healthcare dramatically. Many current problems will become obsolete via new emerging alternatives.

Predictive & Preventive is AI

Both are AI. Period. Providers must employ strong mathematicians and physicists and other scientists to create smarter AI. Google works on duplication of the human brain on non-biological carrier. Qualcomm designs neuro chips. IBM demonstrated brainlike computing. Their new computing architecture is called TrueNorth.

Other healthcare participatory providers [technology companies, ISVs, food and beverage companies, consumer goods companies, pharma and life sciences] must adopt strong AI discipline, because all future solutions will deal with extreme data [even All Data], which is impossible to tame with usual tools. Forget simple business logic of if/else/loop. Get ready for the massive grid computing by AI engines. You might need to recall all math you was taught and multiply it 100x. [In case of poor math background get ready to 1000x efforts]

Education is a Pathway

Both patients and providers must learn genetics, epigenetics, genomics, proteomics, pharmacogenomics. Right now we don’t have enough physicians to translate your voluntarily made DNA analysis [by 23andme] to personal treatment. There are advanced genetic labs that takes your genealogy and markers to calculate the risks of diseases. It should be simpler in the future. And it will go through the education.

Five years is a time frame for the new student to become a new physician. Actually slightly more needed [for residency and fellowship], but we could consider first observable changes in five years from today. You should start learning it all for your own needs right now, because you also must be educated to bring better healthcare to ourselves!

 

Five Sources of Big Data

Some time ago I’ve described how to think when you build solutions from Big Data in the post Six Graphs of Big Data. Today I am going to look in the opposite direction, where Big Data come from? I see distinctive five sources of the data: Transactional, Crowdsourced, Social, Search and Machine. All the details are below.

Transactional Data

This is old good data, most familiar and usual for the geeks and managers. It’s plenty of RBDMSes, running or archived, on-premise and in the cloud. The majority of transactional data belong to corporations because the data was authored/created mainly by businesses. It was a golden era of Oracle and SQL Server (and some others). At some point the RDBMS technology appeared to be incapable of handling more transactional data, thus we got Teradata (and others) to fix the problem. But there was no significant shift for the way we work with those data sources. Data warehouses and analytic cubes are trending, but they were used for years already. Financial systems/modules of the enterprise architectures will continue to rely on transactional data solutions from Oracle or IBM.

Crowdsourced Data

This data source has emerged from the activity rather than from a type of technology. The phenomenon of Wikipedia confirmed that crowdsourcing really works. Much time passed since Wikipedia adoption by the masses… We got other fine data sources built by the crowds, for example, OpenStreetMaps, Flickr, Picasa, Instagram.

Interesting things happen with the rise of personal genetic testing (verifying DNA for millions of known markers via 23andme). This leads to public crowdsourced databases. More samples available, e.g. amateur astronomy. Volunteers do author useful data. The size of the crowdsourced data is increasing.

What differentiates it from transactional/enterprise data? It’s a price. Usually, crowdsourced data is free for use, with one of creative commons licenses. Often, the motivation for the creation of such data set is the digitization of our world or making a free alternative to paid content. With the rise of nanofactories, we will see the growth of 3D models of every physical product. By using crowdsourced models we will print the goods at home (or elsewhere).

Social Data

With the rise of Friendster–>MySpace–>Facebook and then others (Linkedin, Twitter, etc.) we got a new type of data — Social. It should not be mixed for Crowdsourced data, because of completely different nature of it. The social data is the digitization of ourselves as persons and our behavior. Social data is very well complementing the Crowdsourced data. Eventually, there will be digital representation of everyone… So far social profiles are good enough for meaningful use. Social data is dynamic, it is possible to analyze it in real-time. E.g. put Tweets or Facebook posts thru the Google Predictive API to grab emotions. I’m sure everybody intuitively understands this type of data source.

Search Data

This is my favorite. Not obvious for many of you, while really strong data source. Just recall how much do you search on Amazon or eBay? How do you search on Wikis (not messing up with Wikipedia)? Quora gets plenty of search requests. StackOverflow is a good source of search data within Information Technology. There are intranet searches within Confluence and SharePoint. If those search logs are analyzed properly, then it is clear about potential usefulness and business application. E.g. Intention Graph and Interest Graph are related to the search data.

There is a problem with “walled gardens” for search data… This problem is big, bigger than for social data, because public profiles are fully or partially available, while searches are kept behind the walls.

Machine Data

This is also my favorite. In the Internet of Things every physical thing will be connected. New things are designed to be connectable. Old things are got connected via M2M. Consumers adopted wearable technology. I’ve posted about it earlier. Go to Wearable Technology and Wearable Technology, Part II.

The cost of data gathering is decreasing. The cost of wireless data transfer is decreasing. The bandwidth of wireless transfer is increasing dramatically. Fraunhofer and KIT completed 100Gbps transmission. It’s fourteen times faster than the most robust 802.11ac. The moral is — measure everything, just gather data until it becomes Big Data, then analyze it properly and operate proactively. Machine data is probably the most important data source for Big Data during the next years. We will digitize the world and ourselves via devices. Open Street Map got competitors, the fleet of eBees described Matterhorn with million of spatial points. More to expect from machines.

Wearable Technology. Part II

This story is a logical continuation of the previously published Wearable Technology.

Calories and Workouts

Here I will show how two different wearable gadgets complement each other for Quantified Self.  For the beginning we need two devices, one is wearable on yourself, second is wearable by your bike.

First device is called BodyMedia, world’s most precise calories meter. It measures 5,000 data snapshots per minute from galvanic skin response, heat flux, skin temperature and 3-axis accelerometer. You can read more about BodyMedia’s sensors online. BodyMedia uses extensive machine learning to classify your activity as cycling, then measuring calories burned according to the cycling Big Data set used during learning. Check out this paper: Machine Learning and Sensor Fusion for Estimating Continuous Energy Expenditure for excellent description how AI works.

Second device is called Garmin Edge 500, simple and convenient bike computer. It has GPS, barometric altimeter, thermometer, motion detection and more features for workouts. You can read more about Garmin Edge 500 spec online. My gadgets are pictured herein.

On the Route

The route was proposed by Mykola Hlibovych, a distinguished bike addict. So I put my gadgets on and measured it all. Below is info about the route. Summary info such as distance, time, speed, pace, temperature, elevation is provided by Garmin. it tries to guess about the calories too, but it is really poor at that. You should know there is no “silver bullet” and understand what to use for what. Garmin is one of the best GPS trackers, hence don’t try to measure calories with it.

Juxtaposition of elevation vs. speed and temperature vs. elevation is interesting for comparison. Both charts are provided by distance (rather than time). 2D route on the map is pretty standard thing. Garmin uses Bing Maps.

Burning Calories

Let’s look at BodyMedia and redraw Garmin charts of speed, elevation and temperature along the time (instead of distance) and stack them together for comparison/analysis. All three charts are aligned along the horizontal time line. Upper chart is real-time calories burn, measured also in METS. The vertical axis reflects Calories per Minute. Several times I burned at the rate of 11 cal/min with was really hot. The big downtime between 1PM and 2:30PM was a lunch.

An interesting fact is observable on Temperature chart – the Garmin was warm itself and was cooling down to the ambient temperature. After that it starter to record the temperature correctly. Another moment is a small spike in speed during downtime window. It was Zhenia Novytskyy trying my bike to compare with his.

Thorough Analysis

For detailed analysis of the performance on the route there is animated playback. It is published on Garmin Cloud. You just need to have Flash Player. Click this link if WordPress does not render the embedded route player from Garmin Cloud. There is iframe instruction below. You may experience some ads from them I think (because the service is free) …

The Mud

Wearable technology works in different conditions:)