Solutions

Harnessing Big Data for Precision Medicine, Part 2

Applying lessons from the intelligence community to precision medicine

We recognize that to realize the promise of precision medicine, we need an entirely new way of thinking about big data. We must be able to quickly and easily bring together massive amounts of information—unstructured as well as structured—and make it readily available for analysis.

We must be able to let the data talk to us, so we can find those valuable correlations and patterns. We need new ways of ensuring security and privacy, so that the diverse stakeholders of the precision medicine megacommunity will freely share data. And we must be able to put the data and analytics directly into the hands of researchers.

These goals are within reach. A new approach to data and analytics—initially developed through an ongoing collaboration between Booz Allen and the U.S. government to combat terrorism and other threats—is now being applied to precision medicine.

“It is about much more than technology. It really represents a new mindset—a reimagining of how information can be used to further medical discovery.”

It arose from a thorny problem facing intelligence analysts. They were collecting massive amounts of intelligence data from a broad range of sources, and yet had only a limited ability to bring it together. We helped the government develop a new approach, one expressly designed for big data. It has enabled intelligence analysts to use all of their data to see the bigger picture. It has also been used by U.S. military in Afghanistan and Iraq for various intelligence purposes, such as identifying militant groups responsible for roadside bombs.

This new approach has been moving into the larger government and business communities, most notably the health and financial sectors. For example, data scientists at Booz Allen, working with a hospital chain in the Midwest, brought together and analyzed large amounts of data on patients with sepsis, the life-threatening response by the body to infection.

They found previously hidden patterns in thousands of patients’ vital signs, indicating when a patient was about to go into severe sepsis, a more dangerous condition. As a result of the findings, hospitals in the chain immediately started monitoring sepsis patients for such red flags, so that doctors could implement their protocols earlier and save lives.

We believe our new approach is ideally suited to solving the kinds of big-data problems that threaten to hold back progress in precision medicine. It is about much more than technology. It really represents a new mindset—a reimagining of how information can be used to further medical discovery.