'Big Data' may help congestive heart failure patients

According to IBM, we create 2.5 quintillion bytes of data every day. So much that 90% of the data in the world today has been created in the last two years alone.

This data comes from everywhere: from sensors used to gather climate information, posts to social media sites, digital pictures and videos posted online, transaction records of online purchases, and from cell phone GPS signals to name a few.

Computer scientists call this ‘Big Data.’

“The terms big data and data science are gaining significant attention these days, though the idea of making decisions based on large amounts of data has been the dream of many disciplines for a long time,” says Institute of Technology Professor Ankur Teredesai.

Big Data requires far more sophisticated methods of analysis, in part because of the sheer volume of the data and in part because it is often coming in many forms, from many sources and at varying velocities.

The healthcare industry in particular has attracted tremendous attention in recent years as a big data domain where inefficiencies and scale can be extremely challenging and costly.

But an innovative partnership between UW Tacoma and MultiCare Health System, led by Teredesai, will harness technology innovation and data from anonymous electronic medical records to help congestive heart failure patients potentially receive better care at a lower cost and train future data scientists and academics in the rapidly expanding field of ‘Big Data Science.’

The project will look at a wide array of data from anonymous electronic patient records, to search for hidden patterns that may be able to be used to reduce the risk of readmission for congestive heart failure patients to hospital through the emergency departments (EDs). The project will also allow UW Tacoma students to work hands-on with large amounts of data in a health care environment, better preparing them to work in the rapidly growing field of big-data management.

Dr. H. Lester Reed, senior vice president for quality at MultiCare Health System is the project’s senior sponsor and advocate at MultiCare. He believes this collaboration is significant because it may help solve the challenge of finding significant patterns in data from electronic patient records.

“It brings two completely divergent professions around a common cause. When that happens, new concepts are born that neither would have thought of.”

In the long run, Dr. Reed says, the work will help healthcare providers be able to predict the congestive-heart-failure patients who most likely will be readmitted to the hospital and target resources to those patients to help them avoid readmission. This approach will offer better outcomes for individual patients, as well as lower health care costs.

“MultiCare is a leader in using electronic patient records,” says Dr. Reed, but this project couldn’t have been done even three years ago. “The data has become so extensive, but extracting meaning from the data remains a challenge, and that is why the partnership with the UW Tacoma will be so powerful,” he said.

The collaborative research project will engage two graduate students, faculty and research staff working collaboratively with domain experts at MultiCare Health System.

Dr. Reed is also excited that the project will allow hands-on opportunities for students to receive training in working with big data in a healthcare environment.

There remains a high demand overall for technology workers with those kinds of skills, says Christine Roggenbusch, a hiring manager for the startup NPario, Inc. NPario works with big data applications in advertising and publishing and recently hired Institute of Technology graduate Ashish Bindra to work at its Bellevue office.

Roggenbusch says that collaborative projects like the MultiCare big data project are ideal for technology students to gain practical experience, which is very valuable. “When we hire people we want to be able to see that the person has done actual projects in a real environment,” she says

Dr. Senjuti Basu Roy, assistant professor at the Institute of Technology and co-investigator on the grant, says that the project will be “very educational for graduate and undergraduate students and offer a balance between doing research and building systems. The graduate students will have the opportunity to learn cutting edge data science techniques and tools.”

Plus, the project offers a chance to support creating greater diversity in a field that has been traditionally mostly men. “Hopefully this project will attract women students who will contribute to data science.”

In addition to mentoring opportunities, she is excited about the innovation this project offers.

“The exciting and most challenging part is the technology part of the project. We are adept at dealing with a lot of data from heterogeneous sources. The challenge here will be to extract and scale data to create homogenous outcomes and predictions,” she said.

The ambitious eventual goal, she said, is that “the model should be able to predict with high confidence if a patient will be readmitted so we can help mitigate the risk of readmission factors.”