Data Mining and Analytics in Healthcare
With use of the Internet
increasing exponentially, the amount of data being processed has exploded.
Nowadays, even the healthcare business, a world of its own, untouched by the IT
phenomena one decade ago, is forced to embrace the Internet. Currently,
customer behaviors and preferences, their medical information and shopping
patterns are mostly all online. With this amount of information to process, the
need for data mining of this large scale data became more obvious 1.
Data mining, a field at the
intersection of computer science and statistics,1, 2, 3 is the process
that attempts to discover patterns in large data sets. It utilizes methods at the
intersection of artificial intelligence, machine learning, statistics and database systems.1 The overall goal of the data mining process is to extract
information from a data set and transform it into an understandable structure
for further use.2
Doesn’t this sound fantastic? As a private owner of a physical therapy clinic, I would want to know everyone searching the web to
find the right exercises, to minimize joint pain, or to improve their golf
swing after a shoulder surgery. As a director of rehabilitation services, I
would like to find out who is searching the web to find the best short term
rehabilitation center in the area for their loved one. As a physical therapist,
I would like to have software that allows me to enter objective data from my
evaluation and will yield a set of potential diagnoses to optimize my treatment
plan of care. As a mom of two kids who were raised to like organic almond milk,
I would hope local stores are making a note of this fact and are sending me
some coupons for my future milk purchases. That’s data mining and it’s
How can data mining benefit healthcare and how can we embrace this
technological phenomenon? Because standardization of data is at the core of
data mining, healthcare professionals must embrace and perfect the use of EHR.
Dr. Chid Apte is a director of analytics research in the IBM Research Division at the Thomas J.
Watson Research Center in Yorktown Heights, New York. In one of his
publications, Data mining and clinical data repositories: Insights from a 667,000
patient data set, he concluded that data mining technologies “…have the potential
to expand research capabilities through identification of potentially novel
clinical disease associations2.” Using analytics in healthcare is
not only beneficial, but essential. For example, companies such as Edifecs assist in ICD-10
compliance. According to the company, Edifecs ICD-10 Impact Analytics enables
healthcare entities to identify ICD-10 impacts based on their historical data.
This is a critical first step toward understanding the challenge and
determining how to address the ICD-10 mandate. As our healthcare world is striving to become more standardized in
evaluations of diseases, treatment techniques and preventive medicine, data
mining can be the essential step in maximizing clients’ quality of care and
quality of life.
and Web Analytics – Hypothesis Testing: http://www.wmps.com/blog/website-analysis/web-analytics/statistics-and-web-analytics-hypothesis-testing/