IBM Content Analytics can extract unstructured medical data from 50 million documents within minutes for researchers at BJC Healthcare and Washington University.
IBM has
announced that BJC Healthcare and the Washington University School of Medicine
Center for Biometrics will use its
IBM Content Analytics application to help
researchers extract unstructured medical data from 50 million documents.
BJC Healthcare
is a nonprofit health system serving the St. Louis, southern Illinois and
mid-Missouri areas. The WUSM (Washington University School of Medicine) in St.
Louis was founded in 1891.
The two
organizations wanted a way to sort through disjointed and redundant biomedical
data, so that more than a small number of researchers can benefit from the
information, IBM reports.
BJC and WUSM
will use Content Analytics, part of IBM's Enterprise Content Management
business, to analyze the trends and patterns in documents such as clinical
notes, admission/discharge summaries, electronic health records and diagnostic
reports to assist doctors in making diagnoses.
Content
Analytics, formerly called Cognos Content Analytics, uses some of the same
analytics capabilities now found in the
IBM Watson supercomputer, which uses natural
language data-access technology to search for and retrieve information. This
technology, created through a development project called DeepQA, will be used
to answer questions for doctors on health issues such as drug interactions and
test results.
"We think
IBM Content Analytics will be a game-changer in biomedical research and patient
care," Dr. Rakesh Nagarajan, associate professor in the pathology and
immunology department at Washington University, said in a statement. "It
will ultimately accelerate the pace of clinical and translational research
through more rapid and accurate extraction of research-relevant information
from clinical documents."
The software
algorithm can pull information from these documents within minutes, according
to IBM.
"It's the
software and analytics algorithms running on top of that information that
extract out the nuggets for you and surface them into an interface you can go
and explore," Craig Rhinehart, IBM's director of enterprise content
management, told eWEEK.
Still, the
software presents the data from up to 50 million documents in a dashboard view,
an amount of data that people can't process on their own, Rhinehart said.
By being able
to extract key data from up to 50 million documents in medical records, BJC and
WUSM will be able to increase the speed of research, and therefore boost
patient care.
"You can
never read 50 million documents and understand what the trends and patterns were
across 50 million documents; it's impossible," Rhinehart explained.
"You couldn't even take 500 people to do it, because there is never an
efficient way to consistently understand the behavior in those documents and
then figure out all the trends and patterns."
Hypothetical
questions that Content Analytics could analyze include conditions occurring
after a patient stops smoking or the reasons for people tripping on steps and
injuring themselves, according to Rhinehart.
Rather than
taking 10 to 15 minutes to find information on a patient's smoking history,
Content Analytics can take only a few seconds to scan hundreds of documents to
find the answer, IBM reports.
"You
might go explore a set of documents to learn something that would have never
occurred to you to query as a search because you didn't know you were looking
for it," Rhinehart noted.
Meanwhile, IBM
demonstrated a new patient portal platform at the CeBIT IT and
telecommunications conference in Hanover, Germany, March 1-5.
A standards-based platform, the IBM Patient Empowerment
System allows patients to integrate and manage their health care information as
well as receive recommendations or alerts on safer medical treatment. They can
also access third-party health portals, electronic health records, sensors and
home remote-monitoring devices.
Email
Print
