Skip to Content

Publications

Demystifying big data

Conquest - Spring 2013

Insights leveraged to improve outcomes


By Hilary Graham

There’s good news: Massive amounts of useful data are generated every day by cancer researchers around the world.

But there’s also challenging news: Due to the huge amount, its complexity and lack of centralization, much is left unanalyzed.

The consequence: Important insights often remain unknown.

Related story: A fully integrated approach

Take this scenario: A 56-year-old woman enters the clinic with BRAF-mutant melanoma. Her cancer initially responds to available treatments, but then she relapses after a few months. Why did the therapy work initially and then fail?

Big data backed by powerful analytics hold the key to understanding why some patients never respond to treatment, why some patients initially respond and then relapse, and why some patients have long-lasting responses.

To unravel this mystery, specific factors such as family histories, clinical test results and genomic data must be collected from a large number of patients and housed collectively so that factors that modulate response can be identified.

Andrew Futreal, Ph.D. (far left), Keith Perry, Lynda Chin, M.D., and John 
Frenzel, M.D., use a specially designed meeting pod on Floor 1 of 
MD Anderson's South Campus Research Building 4 to discuss next steps 
on the Big Data Analytics Platform.
Photo: Wyatt McSpadden

The big data buzz

Big data is a buzzword today, but what is it and how is it relevant to cancer research and patient care?

“Big data is not just big. The term also implies three additional qualities: multiple varieties of data types, the velocity at which the data is generated, and the volume seen within 
MD Anderson,” says Keith Perry, associate vice president and deputy chief information officer.

“Additionally, many of our databases currently don’t interface with each other because they’re generated by and housed separately in prevention, research and clinical departments versus a centralized platform.”

Harnessing big data is transformative. It provides opportunities to ask complex questions and identify new knowledge in existing data through centralization, standardization and new types of computing analytics.

Harnessing powerful information

MD Anderson refers to the specialized expertise and infrastructure that is being assembled to support the Moon Shots Program as platforms. The Big Data Analytics platform consists of two functional components. First, a longitudinal patient data warehouse will house clinical and genomic data. Second, a suite of massive data analytics will interrogate and learn from the data to provide end users with understandable and actionable answers to their complex clinical or research questions.

“This platform differs from currently available research and clinical databases because of the size, centralization, strength of the analytics and, most important, because it doesn’t belong to one researcher or department,” says Lynda Chin, M.D., chair of the Department of Genomic Medicine and scientific director of the Institute for Applied Cancer Science.

“Providing this wealth of accessible data to all MD Anderson clinicians and researchers as an institutional resource will increase research productivity and improve patient care.”

Enter analytics and their role in a health care setting

Today’s powerful analytics can combine natural language processing of massive amounts of unstructured data with structured research and clinical data. Think doctors’ notes versus numerical lab tests — with sophisticated software to define probabilities when answering complex questions related to medically relevant scenarios.

“Analytics can address large-scale data sets. They permit the discovery of meaningful patterns in data by applying sophisticated algorithms to filter and analyze collected information,” says Andrew Futreal, Ph.D., professor in the Department of Genomic Medicine and co-leader of the Big Data Analytics Platform.

“This approach allows for real-world questions, like the hypothetical scenario above, to be explored because analytics look at interactions and relationships that are not ‘under the lamp post.’ The infrastructure allows for the consideration of all available information about a patient and patients like them.”

Providing the next generation of care today

An interdisciplinary pilot project is under way, providing the next 1,000 patients admitted to MD Anderson’s Leukemia Center the next generation of cancer care. The project integrates scientific and clinical knowledge with a big data infrastructure.

Patients’ information is collected, including a profile of their genetic makeup, clinical histories, test results, treatment courses and treatment responses. This data will be interpreted by the massive data analytics, which provide real-time decision support to rapidly improve clinical outcomes.

This Moon Shots Program platform is a true collaboration between research and clinical departments and information technology.

Democratization of excellent care

The goal of building this infrastructure is to simultaneously harness both technological and human capabilities that will lead to critical insights and dramatically improve patient outcomes.

“Building these integrated platforms enables MD Anderson to leverage the knowledge and insight of scientists and clinicians here to provide decision support, not only to the institution’s physicians, but also potentially to providers across the planet,” says John Frenzel, M.D., chief medical information officer and co-leader of the Big Data Analytics Platform.

“These platforms also could enable physicians globally to practice with the 
MD Anderson standard of care no matter where the patient is located, improving outcomes and reducing suffering.”

Best of all, there will be insights that optimize patient outcomes — as well as the accelerated translation of new therapies and diagnostics — for patients like the woman in the opening scenario and for all patients who are treated with the next generation of medicine.

Request an Appointment

Conquest poll

How much are you affected by gardens or some other healing environments? (select only one)

Make a difference

Your gift to MD Anderson makes a difference in the lives of cancer patients by supporting innovative patient care, research, education and prevention programs. You can Donate Now or learn more at myGiving to MD Anderson.

© 2014 The University of Texas MD Anderson Cancer Center