In This Section      

Managed Care Partners - Mining Big Data to Help Patients

Managed Care Partners Winter 2015

Mining Big Data to Help Patients

Date: January 14, 2015

Mining Big Data
1 2

For the past five years, Johns Hopkins radiation oncologists have used an added tool when forming a treatment plan for head and neck cancer patients: Oncospace. Through this informatics program, conceived by radiation oncology director Theodore DeWeese and medical physicists Todd McNutt and , physicians can plug in a current patient’s information and search a database of other head and neck cancer patients treated at Johns Hopkins with similar tumors to elucidate the best possible radiation plan with the fewest side effects. The program has recently been given to three other medical centers, which are adding in data from their own patients to boost the tool’s repository.

What Kind of Data Is Being Used?

It’s just one example of how Johns Hopkins and other institutions can harness the power of big data—massive amounts of electronic patient information that can be mined to yield individually tailored medical results. Brain scans, genetic codes, family histories, eating habits and medical claim records are just a few examples of information being collected and analyzed.

A university-wide individualized health initiative, known as Hopkins inHealth, seeks to capture this promise. Launched in July 2012, several pilot projects are using big data to improve cystic fibrosis treatment, cancer screenings, scleroderma treatment and interventional cardiology procedures.

“We’re not just treating one patient at a time,” explains Scott Zeger, director of Hopkins inHealth and professor of biostatistics at the Johns Hopkins Bloomberg School of Public Health. “We want to use data from every patient who comes to Johns Hopkins to improve our decisions on every other patient that follows them. We also want to publish the methodology so others know about it, and so the methods and tools can be disseminated more broadly.”

How Does Genomics Factor In?

The initiative could be thought of as a generalization of genomic medicine, where a person’s genes or a set of markers are sequenced and treatment decisions are made based on the results, he says: “Genomics creates one class of data. Our goal is to intelligently use all relevant data, including genomics, to make better decisions for more affordable and better outcomes.”

More projects can be enabled when the High Performance Research Computing Facility opens on the Johns Hopkins Bayview Medical Center campus in spring 2015, a joint project with the University of Maryland, College Park. Funded with $30 million from the state, the center will be equipped with a supercomputer that will be the “most powerful computing system at Johns Hopkins,” Zeger says, with a projected storage capacity of 20 petabytes—the equivalent of 400 million four-drawer filing cabinets full of text.

The pilot projects are “demonstrations of what we can do 1,000 times here at Johns Hopkins for each of our local areas of expertise,” Zeger says. “We can make each of them into a learning health system where as time goes on, we capture data, learn from the data and find new data that will enable us to better care for each of our patients, taking account of his or her individual characteristics, circumstances and preferences.”  

Find Physicians Specializing In...