Overview: data sharing (Scruggs m et al.,


In 2013, The US Institue of Medicine published a
report called “Best Care at Lower Cost”, which described that the insights from
research, and available evidence are poorly used and managed, additionally, the
care experience is poorly captured, which results in missed opportunities, wasted
resources, and potential harm to patients. It called for the development of a “learning
heath-care system” where evidence informs practice, virtuous cycle (Rumsfeld,
et al., 2016). The availability of data, which could help with informing a
learning health care system, has escalated, approaching to zetabytes levels
(10^21 bytes) in the USA for instance (Raghupathi, 2014). The huge volume, and complexity of healthcare
data comes from the diversity of health-related ailments and their
co-morbidities (Dinov, 2016).

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!

order now

Furthermore, Medical big data usually have some
distinctive differences from big data from other fields. They are frequently
hard to get access to, and researchers in the medical field are usually
hesistant to use open data sceince because of data misues by other parties or
lack of data sharing (Scruggs m
et al., 2015). This epidemic increase in the availability of data is anticipated
to continue because of the expansion of electronic health record,
paitient-reported outcomes, data from Internet use, and also genomic
information. However, Big data analytics have been used outside of the
health-care setting by companies such as Amazon or British Airways to improve
sales and effiency (Krumholz, 2014). Therefore, this use of big data has raised hope that Big data
analytics can be applied successfully in health care, with recognition that
target outcomes of other industries are not the same as health-care outcomes (Rumsfeld,

What is big data:

Big Data is data, which are so voluminous and
complicated requiring  a whole new
techniques, algorithms, and anlytics to deal with them. Big data are usually
defined by the four ‘V s’, volume, variety, velocity, and veracity (Bellazzi,
2014). The Volume is the quantity of data generated and stored. Their volume
does not have a standard definition, but most sets contain at least 1 petabyte,
which is 10^15 bytes (Rumsfeld, et al., 2016). Big Data has the potential of
representing the real world without bias, linking with other datasets, being
reused, accumulating value over time, and innovating a multi-dimensional,
system-level understanding(Scruggs, et al., 2015). Analysis
of the data can find new interrelationship to “spot business
trends, prevent diseases, combat crime and so on” (The
Economist, 2010). However, with the large volume of data, the
information big data provide sometimes might be unsuitable for what the researcher intent to (Sinha, et al., 2009) . They
cannot be expected to be treated as individual datasets, so that to add value,
they need to be linked with other potential data sets (Sinha, et al., 2009).

Data Science promise for supporting Cardiovascular

Data Science, or Data-driven Science, can be described
as the methods, processes, systems of extracting, inferring, and validating
knowledge or insights from data sets in various form. It creates tools and reinforce
access of datasets for researchers. The vision for Big Data science to benefit
the cardiovascular community in delivering better treatment. The potential for
Big Data science to improve cardiovascular quality of care and patient outcomes is tremendous.
promise of Big Data lies in the “insights generated from these databases and
analyses, which have the power to inform and improve health-care delivery and patient out-comes” (Rumsfeld, et al., 2016).  This potential uses to improve cardiovascular
investigations, cares, and paitent outcomes will come from individual patient care, therapeutic
decisions, so that it will guide the effective uses of resources in health-care
systems, and in plublic health applications. However, these data are in innumerable and non-commensurate state, which prohibit the interoperability of
the investigators. For instance,