Biomedical Engineering Reference
In-Depth Information
20
Extreme scale clinical analytics
with open source software
Kirk Elder and Brian Ellenberger
Abstract: Knowledge is at the root of understanding all symptoms,
diagnosing every ailment, and curing every disease. This knowledge
comes from the deep studies performed by research organizations
and diligent healthcare workers who contribute to documenting and
responsibly sharing their observations. Through the American
Recovery and Reinvestment Act of 2009 (ARRA [1]), the industry
was incented to implement electronic medical record systems that
capture more information than ever before. When billions of medical
records converge within a secure network, the baton will be handed
to analytics systems to make use of the data; are they ready? This
chapter explores what the next-generation software infrastructure
for clinical analytics looks like. We discuss integration frameworks,
workfl ow pipelines, and 'Big Data' storage and processing solutions
such as NoSQL and Hadoop, and conclude with a vision of how
clinical analytics must evolve if it is to handle the recent explosion
in human health data.
￿ ￿ ￿ ￿ ￿
Key words: clinical data; ICD10; HL7; electronic health records; Big
Data, NoSQL.
20.1 Introduction
One of the largest problems in clinical analytics is that the immense
breadth of available services results in a very diverse set of implementation
details. Standardizing every scenario quickly becomes impossible. The
 
Search WWH ::




Custom Search