A Conversation about Big Data
Euan Ashley, MD, Chair of the Biomedical Data Science Initiative
December 2014
With Euan Ashley, MD, Chair of the Biomedical Data Science Initiative
What is the Stanford Biomedical Data Science Initiative?
The Biomedical Data Science Initiative (BDSI) is a School of Medicine initiative, which brings together faculty members with an interest in computation and data science and its applications to medicine and human health. Our tagline is "Harnessing the world's data to improve human health." We are interested in large data sets on a population scale, but also applying those kinds of insights to the personalized medical care of individuals. We have already given out almost a million dollars in seed funding for 12 different grants. These seed grants will fund science using population scale data, data from developed and developing worlds, mobile health, personalized medicine and genomics, proteomics and immunology.
Where is big data affecting medical care of our patients today?
The application of large-scale genome sequencing is one of the few places where big data is beginning to impact our patients directly. The Clinical Genomics Service at Stanford, which is in its pilot phase, is bringing genome sequencing to the bedside of both our adult and child patients.
How will big data affect patient care in the future?
Our strong belief is that the kind of technology available today, the same technology that powers Google and Amazon, can absolutely transform medicine. You can go to Google and search an enormous database of the world's information in microseconds from any browser anywhere. The question is: why can't we do that for medical information?
At the moment, we can't conduct these types of searches because the medical data is hidden and siloed behind firewalls that are cultural, ethical and legal. One of our dreams for this initiative is to break down those walls and connect the world's health care data in a way that would allow us to essentially "Google" the health care data of millions or even billions of people to help the patient sitting in front of us.
What obstacles do you face in mining big data?
The technology companies that live just outside our door have solved these problems already—taking large data sets, searching them very fast and doing computational modeling on complicated data sets. That's why we're excited about this and feel that it can be done. The challenge is getting the data, and getting it into a form that can be searched and made available for analysis. And then scaling that globally.
How is Stanford positioned to advance the science of big data research?
I think Stanford is better positioned than any other place on the planet to do this. There are a number of reasons for that. We have a phenomenal background in all the sciences that are required to do this. We have the number one computer science department. We are very strong in statistics and mathematics. We have one of the best bioinformatics programs in the country. We're number one in genetics. Plus we have great collaboration between all of those groups. In just ten minutes, you can walk from the engineering school to the medical school to the hospital. I think the proximity of these departments is extremely beneficial.
And we're in the middle of Silicon Valley—we are surrounded by the very companies that have already changed the world. These companies are our advisors, partners, collaborators and colleagues. The Biomedical Data Science Initiative really stretches across departments and across schools and into Silicon Valley.
What else is the BDSI focusing on for the future?
We are interested in modeling biological systems—something that requires large computation. From cellular models, we try to work out how individual cells work and how they interact in organs around the body. We're interested in how you can use search data to monitor things like flu or other infectious disease outbreaks and how such epidemics can be mapped via knowledge of the genomics of the bug that's causing it. We can sequence the bacteria that cause infectious disease and track that along with tracking the people infected. This will impact infectious disease in a major way.
Another major area of focus in the coming year will be mobile health. All of a sudden, we have smart watches that can measure heart rate and activity and sleep, and mobile phones that incorporate health care data in a mobile format. We will put these new generation devices onto patients and start to see what kind of effect having that kind of information can have on health and well being.
Is the next big medical discovery already out there? Does it just need to be mined from data that already exists?
It is perfectly plausible that the next big discovery will come from data that already exists. Data is available to anyone in the world with a computer and a browser and that's the exciting thing now. The number of data sets that are publicly available for anyone, even a high school student, to go and analyze is simply astounding. There are very few areas where our ability to manipulate big data would not essentially transform the way we do medicine.