Rong Xu Searches for Medical Breakthroughs in Big Data
A tenured Associate Professor in the Department of Epidemiology and Biostatistics at Case Western Reserve University, Rong Xu (Stanford, CS ’04) analyzes data to find patterns that reveal how a drug (or a combination of drugs) might be used to address a different disease than what the drug was originally designed for.
Medical research usually evokes images of white lab coats, microscopes and sterile rooms, but the next big breakthrough in healthcare may come from algorithms that analyze vast amounts of highly complex and heterogeneous biomedical and clinical data.
At least that’s what Siebel Scholar Rong Xu (Stanford, CS ’04) is hoping as she leads a group of students in the emerging field of biomedical informatics, the analysis of scientific data by researchers in biology, public health and other health sciences. Xu is a tenured Associate Professor in the Department of Epidemiology and Biostatistics, School of Medicine, Case Western Reserve University, where she analyzes data to both to reveal the mechanisms that underlie human diseases and discover new drug treatments to combat them.
When medical drugs are in their research, development, clinical trial and post-marketing phases, researchers and consumers document a broad range of intended and unintended side effects. Using natural language processing and machine learning algorithms that they develop, Xu and her team search through that data looking for insights and patterns of how a drug, or a combination of drugs, might be used to address a different disease than what the drug was originally designed for. While biomedical informatics research has not yet led to a medical breakthrough, Xu’s work has been recognized by her peers because it shows great promise as an innovative pathway to new drug treatments. In 2015, she was awarded the Landon Foundation AACR INNOVATOR Award for Cancer Prevention Research and the American Medical Informatics Association (AMIA) New Investigator Award. In 2014, she was recognized with the NIH Director’s New Innovator Award. Rong was featured on 2016 Cleveland’s Crain’s “Who to Watch Health Care,” which highlights innovators in Northeast Ohio’s medical fields
Xu, who received both her Ph.D. in Biomedical Informatics, M.S in Computer Science, and Certificate in Graduate School of Business Summer Institute for Entrepreneurship, from Stanford University, and B.S in Biology from Peking University, spoke with the Siebel Scholars program about her research.
Q: What inspired you to work at the intersection of computer engineering and biomedical sciences?
I have studied both biology and computer science, a somewhat rare combination, and I wanted to find a way to combine the disciplines. Biology (working with live organisms) is the traditional path to medical research and pharmaceuticals, while data analytics have been used by many different industries for years. There are layers of data that document the side effects of drugs during both R&D (research and development) phase and post-marketing consumer phase. I thought it would be interesting to cull through that data to see if drugs being developed for a specific disease could be used to for a completely different diagnosis. Perhaps the best-known use of this was when scientists discovered that infamous ‘little blue pill’ developed for heart disease could help with erectile dysfunction.
Q: How would you describe your work in simple terms?
I develop computational algorithms, including natural language processing, machine learning, data mining, and artificial intelligence, to discover potential solutions to diseases that have no cure like Alzheimer’s, Parkinson’s, schizophrenia, cancers and many other common complex and rare genetic diseases. We use computational techniques that can create, integrate, and analyze large amounts of heterogeneous and complex biological and health data to look at range of variables such as patient genetics, age, gender and location, as well as side effects from various drug research.
Using patient electronic health records (EHRs), we develop computer algorithms to design large-scale ‘in silico’ case-control and cohort studies to rapidly analyze existing clinical data in a new way, searching for patterns that maybe different than the original purpose of clinical uses. For example, the regular use of aspirin to reduce the risk of heart attack or stroke may have another important benefit in decreasing the risk of some types of cancer. Our own study shows that certain FDA-approved nonsteroidal anti-inflammatory drugs can robustly kill ovarian cancer cells.
Q: What projects are you current working on?
Working with my students at Case Western Reserve University, we create computational algorithms to see if there is anything that can be extracted to address health problems like ovarian cancer or mental disorders. We allow data lead us to new discoveries instead of having a specific hypothesis and using data to verify it.
Q: What are some of the challenges on these projects?
Any work in the sciences requires patience. A researcher can spend hours or even years in a lab and not yield their desired results. This is also true when working with biomedical informatics because it is hard to prove that what you find is novel and unexpected. We look for patterns and commonalities before we can test a specific theory. Eventually, any findings would be shared with, and verified by, biomedical researchers. Hopefully our computational research can streamline the process.
Q: What is most rewarding to working in the biomedical informatics field?
Biomedical informatics is a relatively new field and it has been great to see students embrace the possibilities of this type of work. It is also rewarding to see the academic community and medical institutions see the potential of this type of research.
Q: How has being a Siebel Scholar impacted your work?
Being a Siebel Scholar gave me confidence to pursue my Ph.D. in Biomedical Informatics from Stanford University. It has encouraged me to take risks to pursue this emerging field and helps bring credibility to the scientific merit for this type of research.
Q: What advice would give to your students or others about pursuing pioneering work?
You must have passion for the work you do. Working in a relatively new field like biomedical informatics, you must have a larger motivation. For me, it is the idea that our work can impact public health challenges like cancer or Alzheimer’s disease. Be prepared for hard work, with high risk factors and, hopefully, high reward and high impact.