“Big Data” is justifiably one of the most exciting new tools available to various industries today, including medical science. Given that a portion of the work we do at Rock West Solutions is focused on the healthcare sector, we have a front row seat as big data transforms the way we research, develop, and deliver medicine. While it is easy to recognize the inherent potential of this field, it is another challenge to discover how to make most effective use of it. One thing is clear: utilizing big data in medicine is all about asking questions.
This may sound counter-intuitive at first. Wasn’t the entire big data concept originally conceived as a way to provide answers? Consider briefly the traditional approach of scientific inquiry. A researcher would identify a specific question they wanted answered, and then design an experiment to gather data pertinent to answering that particular question. This still happens to some degree today, but the technique is becoming more obsolete as it becomes easier to gather more broad datasets. Why only gather data specifically for one question when, using the same amount of resources, you can cast a much larger net and possibly answer dozens of questions? There is no shortage of data, now the key becomes figuring out the right questions to ask.
It’s All About the Questions
Big data was first introduced as a concept that called for finding ways to gather and store as much data as possible. The thought was that once it was gathered, one could go back and find ways to use it. The problem is that our ability to collect data has far outstripped our ability to find value from it. We are collecting data so fast that we simply cannot keep up with it.
Atul Butte, M.D., Ph.D., chief data scientist for University of California Health, contends that it doesn’t have to be this way, especially in the field of medicine. He told an audience during a recent speech in San Diego that the inability to use big data to its full potential in scientific research is obviously not due to a limited data set, it is due to not asking the right questions to fully exploit the power of this data. Butte, among others, was presenting in regards to the University of California Health Data Warehouse. This revolutionary initiative is a clinical data repository for all six University of California health systems, bringing together the medical histories of every patient from all facilities.
A central data source of this volume is an exceptional resource for the medical researchers in these institutions. “Historically, I designed a study, conducted the study and looked at the data generated by that study,” said Paul Mills, Ph.D., a professor in family medicine and public health at UC San Diego. “Now I can start by looking at what has happened clinically to patients in our existing database. What I have learned today is that I can access a lot of the health data myself and get myself to a point where the next step is IRB approval to go deeper into the data.” By asking the right question, Mills is able to save time and resources usually necessary to begin a study, and is able to begin his research with a more accurate idea of how a condition manifests in the real world.
Pull Rather Than Push
Collecting data in mass and then analyzing it in its entirety to see what useful information can be gleaned relies on the data set ‘pushing out’ relevant information. The ‘push’ method may work in some settings, but it is not viable in the big data arena for one simple reason: volume. There is just too much information to push and still reach concise, viable conclusions. It’s better to pull out the information you need. And to do that, you need to ask questions.
Mills method of asking specific questions that can be answered by the data rather than poring over data to find out what it can teach him makes much better use of access to a massive dataset. Imagine having a set of encyclopedias. Does it make more sense to read through every volume and hope to come across relevant information or to use the index to acquire knowledge on one specific subject? Asking questions forces us to focus on the data as a means of finding specific information necessary to satisfy the topic at hand. Pinpointing the inquiry makes it possible to navigate the tremendous volume of data and actually find something of use.
Rock West Solutions can work with a customer in the healthcare sector to develop better software and signal processing methodologies, but more than that, we can help make more efficient use of data being collected. Even the most massive dataset is only as good as the analysis tools used to process it. Rock West has successfully applied statistical methods to process thousands of data sets to answer important questions for our customers. Boasting highly experienced team members with a variety of complementary skillsets such as statistical analysis, requirements analysis and management, signal processing, and optimization algorithm development, we are able to adapt to any project. Big data is the way of the future, and Rock West Solutions is committed to helping our customers take full advantage of the power of this new paradigm.
- University of California – https://www.universityofcalifornia.edu/news/big-data-big-wins-medicine