Big Data in Healthcare (Part 2)

Big Data is Watching You

The medical industry provides an enormous data flow, which if managed correctly can enable a more empowered engagement of medical professional and the pharmaceutical industry with the general populace at large. The information gathered on all types of individuals with all kinds of complicated conditions or illnesses can be dealt with in a more intuitive and holistic manner using analysis drawn from the huge amount of data out there already. Digitisation of this information provides big data analytics with a chance to provide systems enabling, cross-party diagnostics, accountability, detection of disease and tracking of pandemics. It allows for managing sections of the population prone to certain conditions, managing cost, detecting fraud, predicting outcomes, estimating duration times in hospitals. It can identify risk factors for certain patients, anticipating patient complications or their susceptibility to hospital-borne disease like MRSA, etc.

A report from 2013 entitled the big-data revolution in healthcare mapped out a new paradigm which would service all stakeholders but holding to four central tenets i.e.

  • Right Living – Proactive informed engagement and good health management.
  • Right Care – Correct healthcare package and facilities for a clinical impact.
  • Right Value – Consumer decision making us reduce cost increase value.
  • Right innovation – ‘To advance the frontiers of medicine and boost R&D productivity in discovery, development, and safety Ecosystem feedback.

The report goes on to say in the US a concerted effort by governments to encourage the use of EMRs (Electronic Medical Records) has provided a dividend in terms of collectible data. Since 2011 over fifty per cent of Doctors in the US have been using EMRs and over seventy-five perc cent of Hospitals. Also, the advanced software systems version EHRs (Electronic Health Records) provide a more detailed history of the patient. They allow patient health records move with them and are accessible to all medical professionals involved with the patient, no matter where they may be receiving treatment. They can cover the patient’s total health history including MMR, X-rays, Doctors reports, diagnostic history, medicines prescribed, and so forth. This allows data tracking, monitoring and improving the care of patients with a tailored approach to their particular needs. (Mc Kinsey, 2013)

The use of Big Data is not limited to the USA, Britain’s NHS is considering rolling out a number of wearable monitors while in Ireland in 2010 the HSE,  instigated the NIMIS project spending over €40m on rolling out a state of the art electronic radiology systems for 35 Irish hospitals. It made X-Rays and CT Scans digital and rapidly transferable throughout health services. Now patient’s records are easily accessible to all medical practitioners involved with the patient.  Services such as this add to the data stream of the healthcare sector and will provide the clinical intervention models for the future.

Medicine, Wearable Technology and the Internet of Everything

As previously mentioned, the medical sector is constantly creating masses of data both online and offline, which if collected and managed correctly, it can lead to a more personally tailored medical response for the patient. Medical professionals should be able to accurately diagnose and predict present and future medical scenarios as indicated by previous lifestyle choices, eating patterns, BMI, and the various other empirical indicators to future plan for a person’s health. Big data can provide us with analysis of who, what, where, when, and how a patient has arrived at a certain place in time. There are many factors of a person’s background which will provide indicators to their possible susceptibility to disease and preventative measure they can take to avoid it.

The phenomenal growth of advanced diagnostic tools, wearable technology to monitor health and patient profiling will lead to a revelation in medicine over the next ten years. With the aid of smart devices such as wearable body monitors, big data analytics can be used to radically increase the patient involvement and interaction with their treatment and ongoing communication with professionals who may be located remotely at some distance from the person. Imagine being able to access a world class consultant in another city or country via your smart phone, or having their opinion of your condition sent to your local Doctor. This is just one possibility with this technology, this is the technology of the Internet of Things. The diagram from Cisco estimates the number of things connected to increase significantly, to become the internet of everything. Many will have our health and well-being as their core function.

Diagram 2: Source: Cisco

Smart Medical Wearable Devices

We are all familiar with pulse rate counters and heart rate monitors used in the gym or by runners. We are also familiar with the idea of nicotine patches which help smokers quit. The basic principles of these technologies has been expanded and developed into a multi-billion dollar industry worldwide with companies such as Sano Intelligence with their breakthrough sensor and software technology using their expertise in biochemistry, data science, hardware, software and semiconductor to produce a ‘biometric sensor that will help people understand what’s happening inside their bodies’. (Sano 2015)

There are other patches which will monitor vital signs and blood pressure. These are unobtrusive devices with wireless communication technology monitoring people at risk from their condition and have the ability to improve the lifestyle and medical intervention of millions of people over the next few years. From its own mobile health wellness sensor reports ON World research predict that by 2017, over 500 million wearable, implants, mobile health devices will be sold worldwide.  (Mobihealthnews, 2013)

ON World predicts that longer life blood glucose implants will benefit diabetics and lower treatment cost. They also state that they see wearable sensor technology as the largest growth area increasing by 552%. This includes smart watches which will be used increasingly for health and fitness monitoring. This is the area of preventative medicine although also used increasingly as a fitness benchmarking for people who strive ever more to be as fit as they possibly can. Clothing with sensors impregnated into the cloth is also an example of this lifestyle trend. Another very exciting piece of news is that of an iTBra which could be used to check for abnormalities in the breast. Developed by scientist Rob Royea, and backed by Cisco, the sensor bra contains patches which gauge small temperature fluctuations in the tissue of the breast which can be an indicator of the disease. (Cisco, 2015)

Other useful medical wearable technology is related to the neurosciences, dealing with brain activity, inactivity or illness due to stroke accident or disease. Smaller unobtrusive monitors for brain activity are now available and they can read a wealth of information which may be useful for medical purposes. EEG can use various technologies available at present, including face coding and emotive software and eye tracking. Used together with other wearable monitors such as heart rate monitors it may be possible to build up a picture of a patient’s emotional response to a stimulus. This would be useful in patients with diminished cognitive function.

As well as the major health conditions devices are coming on stream for such diverse conditions as Asthma to Ulcers. Googles has developed smart contact lenses for diabetes sufferers or just those who need help reading. And virtual doctor visits are also now possible using wearable technology monitoring devices and video calling, thus avoiding a risk of cross infection for vulnerable patients.

So now we have the data, what do we do with it and how do we do it?

Workable Frameworks

Many companies have come up with solutions to analyse large amounts of data successfully and with the volume of the data set to increase this is an ongoing challenge. Healthcare data can come from anywhere including other computers, health monitors, social media, healthcare profession,  the pharmaceutical industry, insurance companies, billing data, biometric health records, documentation from doctors etc. All this data has to be collected and processed, a framework has to be put in place to make the data useful. There are many ways to analyse, aggregate, control and visualise the data at this stage.  One popular platform is Hadloop Apache, which is a data organiser and analyser. It is a NoSQL type of technology which evolved from SQL. If your data is not changing in structure and has low to moderate growth then SQL should match a company’s needs.

But for data such as healthcare which is in a constant state of flux, the NoSQL is more suitable. (Dataconomy, 2014)

The title NoSQL includes numerous types of database hosts all with different storage models.  This allows NoSQL, the ability to analyse large amounts of data with algorithms using several servers each individually working on a separate area, then bringing the outcomes to the central area to give the end result. It uses horizontal scaling which means, using many servers which can be in the cloud or in-house hardware. This makes it cheaper than SQL which uses vertical scaling which needs huge storage and this can be expensive.

However, Hadloop can be hard to understand for those who have to manage the data such as hospital administrators. There are various companies supplying their platform solutions such as IBM, Cloudera, Hortonworks and there are cloud versions available which make them easily accessible to most organisations. It is inevitable that Big Data analytics is the future of the healthcare sector and over time it will become a standard practice among healthcare professional to use these tools to better practice their profession. ‘To succeed, big data analytics in healthcare needs to be packaged so it is menu-driven, user-friendly and transparent. Real-time big data analytics is a key requirement in healthcare.’ (Raghupathi, 2014) For general uptake user-friendly menus such as drop downs will need to be designed. So a friendly user interface is essential if big data is to taken up by the general medical establishments and used effectively. Benefits should be highlighted and it will require upskilling and training, but the results should speak for themselves.

Conclusion

As with all new tools there will be resistance but this in time will dissipate and when the wrinkles are ironed out in the systems Big Data will provide an additional and very powerful tool in the fight against disease and improving the health of the world’s population. There is confidentiality issue of course but it is curious that people will share information with a commercial company easily but be reluctant to do so with medical providers. This, of course, can be for insurance premium reason and this needs to be taken on board when collecting patient data.  The predictive intelligence that big data produces is already available in certain healthcare systems like the NHS in Britain. But there is still mistrust from the public, even though the potential for it to assist in treatment is clear. Also, medical practitioners themselves are sceptical of taking information as true if they have not gathered it themselves. This may also be because of the risk of legal action if they act on information without being responsible for its collection. There is also concern that insurance businesses and medical administrations could put cost saving before actual patients medical outcome and use big data science to prove their case, in a worst case scenario denying treatment as a negative outcome is definite.

It all depends on whether you are wearing your utopian or dystopian glasses with regard to advanced technology and its impact on society. What cannot be denied is the ability of big data to improve medicine and healthcare for all humanity. But as with all technical advancements throughout history, there will no doubt be downsides. With this in mind, we must as always progress with care.

 

Bibliography

McAfee, A., Brynjolfsson, E., Davenport, T. H., Patil, D. J., & Barton, D. (2012). Big data. The management revolution. Harvard Bus Rev90(10), 61-67.

Transforming Health Care through Big Data Strategies for leveraging big data in the health care industry [Online] http://ihealthtran.com/wordpress/2013/03/iht%C2%B2-releases-big-data-research-report-download-today/ (Accessed 20/10/2015)

Sano is a biometric sensor and software company with a patented, breakthrough technology [Online] http://www.sano.co/ (Accessed 27/10/15)

Prediction: Wearables to lead the 515 million sensors to ship in 2017 [Online]         http://mobihealthnews.com/22752/prediction-wearables-to-lead-the-515-million-sensors-to-ship-in-2017/ (Accessed 27/10/15)

The ‘big data’ revolution in US healthcare [Online]   http://healthcare.mckinsey.com/big-data-revolution-us-healthcare (Accessed 27/10/15)

Using Health IT to Detect Breast Cancer [Online] http://blogs.cisco.com/healthcare (Accessed 27/10/15)

SQL VS. NOSQL- WHAT YOU NEED TO KNOW  [Online] http://dataconomy.com/sql-vs-nosql-need-know/ (Accessed 29/10/15)

(Raghupathi, 2014) The Internet of Things, Wearable Computing, Objective Metrics, and the Quantified. Journal of Sensor and Actuator Networks ISSN 2224-2708.  [Online]      http://www.hissjournal.com/content/2/1/3

 

Appendices

Diagram 1 IBM [Online]     http://www-01.ibm.com/software/data/bigdata/  (Accessed 27/10/15)

Source: http://www.sciencedaily.com/releases/2012/07/120712224622.htm (Accessed 12/12/16)

Diagram 2 CISCO [Online] http://blogs.cisco.com/diversity/the-internet-of-things-infographic   (Accessed 27/10/15)

 

 

css.php