The 20th century was an exponential growth period for healthcare, making the leap from middle age suppositions and methods to modern methods, all the way to genomics.
Due to technological advancements, IoT, and Big Data, the 21st century will probably be the moment when man and machine will work together, and diagnosis will be no longer be just a matter of medical skill, but a combination of the results of human reasoning and deep learning.
In an industry estimated to be spending about $3 trillion a year, and set to account for approximately 18% of the American GDP by 2018 – there is room for improvement and cost savings along with efficiency boosting. The target is to use analytics in a predictive manner since after all, prevention is better than a cure.
Types of healthcare data
Big Data is a mix of various types of information which, by size, velocity or diversity can’t be analyzed and used by employing simple statistical methods. The healthcare sector is by default a generator of such data in the form of patient records, results of lab tests, medical research data, records from sensors and fitness trackers and CRM systems of hospitals and clinics. Each of these types of data can inspire applications of Big Data in healthcare. Here are the seven ways Big Data can be a game changer in healthcare.
Electronic health records (EHR) data mining
The US has already implemented this, while the EU is still lagging. Not only is this an eco-friendly measure reducing the amount of wasted paper drastically, but EHR is also fast and profitable. Doctors can get access to a patient’s status on a tablet or smartphone immediately, make connections to past conditions and keep track of treatment even when the patient changes medical units.
Even more important is the ability to use these records as a primary source for deep learning algorithms. The neural networks can take as input the data stored in millions of files and scan through patterns to uncover underlying connections and help early diagnosis.
Real-Time monitorization
With a steady rise of IoT that can be used to track a wide array of vital signs, there is a decreasing need for medical visits and medical attendance in wards. The morning and night rounds of doctors and nurses can be radically improved by turning them into continuous monitoring and on-demand intervention for patients who experience problems and are in threatening situations. This can even help by freeing up some hospital beds and continuing to monitor the patients at home, for a longer time.
Big data consulting firm InData Labs recommends either batch processing of repetitive information that comes in large volumes or deployment of individual analysis methods for multi-structured data that requires a response on the spot, like acting based on a patient’s vital signs. Even a live feed stream can be analyzed almost instantaneously.
All the data gathered in this process can be further studied to create recommendations related to shortening or extending the medication period, possible complications and co-morbidities.
Predictive analysis
Since what Big Data does best is to identify similarities and patterns, it can be used to make sense of lab tests and medical imagery to understand pathology better and see warning signs before it is too late for effective treatment.
One of the high-profile applications of this is trying to find a cure for cancer or at least sounding the alarm early enough to avoid uncontrollable growth of malign cells. Tumor samples or photos of skin patches can act as training data for deep learning to find mutations that could lead to cancer.
In the US and Canada, opioid medication abuse is a national problem. Overdoses can be lethal, and it is so common that a program had to be put in place. Preventing drug abuse can be done by looking at the patient’s behavioral patterns, the frequency of visits or even the area they live and other factors such as a stressful job or recent surgery interventions.
Health risks management
What if at some point you could receive texts related to your health status? Simple reminders like those might be received from your fitness tracker, but can also be adapted to a far wider range of needs and diseases. Weaving in Big Data into a personalized care plan is a way medicine will become more patient oriented. It’s a paradigm shift from the costly approach of applying a standardized set of tests and sorting out the condition of the patient. Now the struggle is to start with the patient’s values and identify the risks before they turn into disease.
Hospital efficiency
Predictive analysis can be used to anticipate staffing and medication needs and be prepared for seasonal outbreaks such as the cold and influenza in the winter. Linking the inventory to a real-time data processing algorithm can be used to trigger low-level alerts and generate automated replenishment orders. This just in time approach suggested by the Six Sigma and lean methodology is used in production to reduce inventory costs but can be successfully adapted to a hospital environment.
Telemedicine
Different forms of telemedicine were created in the last half century, but the current state of technology allows healthcare professionals to offer their expertise regardless of geographical distances. By incorporating sensors and almost real-time data flows, doctors can even attempt remote surgery with the help of robots. This is a great opportunity especially for secluded areas which don’t attract top medical talent.
Obstacles to implementing Big Data
To be highly efficient, Big Data needs to be cross-analyzed to find the right connections. Since there is a high fragmentation of the storage media, access rights and types of data, there is still room for improvement. A good idea includes integration of these sources in a unitary dashboard from which information can be easily extracted. As a recent Fortune article puts it, the problem is not Big Data, but messy data.
These examples of Big Data use in healthcare are just scratching the surface related to what can be done for improving outcomes in the medical act. The future target is to reduce hospital visits, redirect funding and medical resources towards research and provide each patient with enough information to keep them healthy instead of treating them.
The views, opinions and positions expressed within these guest posts are those of the author alone and do not represent those of Becker's Hospital Review/Becker's Healthcare. The accuracy, completeness and validity of any statements made within this article are not guaranteed. We accept no liability for any errors, omissions or representations. The copyright of this content belongs to the author and any liability with regards to infringement of intellectual property rights remains with them.