Data Science

Data Science at the DCRI

Recent innovations in data collection methods mean researchers have access to deeper and more complex health information than ever before. Even greater opportunities are on the horizon. But clinical trialists often struggle with how to properly frame research questions that take advantage of all “big data” offers.

 
The DCRI Analytics and Data Science team is a trailblazer in creating critical data science methodologies—crafting queries to inform care, correcting for anticipated statistical challenges, and harnessing the power of machine learning to advance research. By tapping the experience of Duke faculty statisticians and the real-world insights of practicing physicians, we find the question that most effectively probes the challenge in your research and leverages the appropriate data to its fullest potential. 

 

The right questions. The right data. Right now.

Big Data, Big Mess

Michael PencinaWhat do clinical researchers do when they find themselves awash in a flood of data? How do they make sense of data from multiple sources? In an editorial in STAT, DCRI Faculty Associate Director Michael Pencina, PhD, weighs in on the challenges presented by the new age of data.

Seeing the Whole Elephant: Potential and Pitfalls in Big Data for Clinical Research

Former FDA Director Rob Califf, MD, Vice Chancellor for Health Data Science, Duke Health and Michael Pencina, PhD, DCRI Faculty Associate Director, discuss the potential for big data in applications for clinical research.

 

 

How We Make Analytics Meaningful

MeaningfulAnalytics

Dam Data: Health Systems, Machines, And Learning

Machine learning allows practitioners to reach beyond their own experiences and access the sum of experiences of all patients in the health system.  It mimics how physicians think about treatment options, but at a scale that can only be achieved with computing.

But machine learning is not sufficient on its own.  Clinical expertise, study design, and deep understanding of the data to be analyzed are as critical as the machine learning methods to be applied.

 

Insights from the Team

Join Us at Upcoming Events

ISPOR 2018
May 19-23 in Baltimore, MD
Booth #432

Lisa Wruck on "In the Electronic Health Record Era, Do We Still Need Clinical Registries?"

 

ATS 2018
May 23 in San Diego, CA

Kevin Anstrom on "Novel IPF Trial Design in the Era of Anti-Fibrotic Therapy"

Data Across Duke

What is Data Science and How is it Changing Medical Research?

In this podcast, Mary E. Klotman, MD, dean of the Duke School of Medicine, discusses how the new Health Data Science Center will bring together quantitative scientists and experts in machine learning to address the most important questions facing clinical researchers.

Lesley Curtis
Lesley Curtis
Chair, Duke Dept of Population Health

The Duke Department of Population Health Sciences engages faculty members from a variety of disciplines — including epidemiology, health services research and policy, health economics, health measurement and behavior, and implementation science — who work to answer complex questions about the drivers of health in large populations.

The Forge is Duke University’s center for health data science. Based in the Duke University School of Medicine and led by Vice Chancellor for Health Data Science Dr. Robert M. Califf,  the Forge is part of a community spanning the university, the Duke health system, and beyond that is using data science to develop actionable insights and improve health outcomes.

Learn more about how the Forge and the DCRI collaborate on the Forge blog.

Duke-Margolis brings together capabilities that generate and analyze evidence across the spectrum of policy to practice, supporting the triple aim of health care – improving the experience of care, the health of populations and reducing the per capita cost.

Learn about Duke-Margolis' proposal for regulatory use of real-world evidence.

Can Randomized Pragmatic Trials Fulfill Their Promise?

Michael Pencina, PhD, Eric Peterson, MD, and co-authors from McKinsey & Company, describe the steps that researchers, the private sector and regulators can take to design and implement clinical trials that are better, faster, and cheaper using real world evidence.

F94A3692CEM Meeting Terrence Jones Photography

AHA and DCRI partner to accelerate precision medicine through machine learning

A new strategic alliance will target the prediction, prevention and treatment of cardiovascular diseases using artificial intelligence computing and big data, the American Heart Association (AHA) and the DCRI announced today.

The AHA’s Institute for Precision Cardiovascular MedicineTM together with the DCRI’s data science team, under the direction of Michael Pencina, PhD, and Lawrence Carin, PhD, will develop and test machine learning methods on the AHA Precision Medicine Platform, which is powered by Amazon Web Services.

Datavant and DCRI partner to accelerate innovation in data-driven clinical research

Datavant, a healthcare technology company that aggregates and structures healthcare data to generate actionable insights for clinical trials, has selected the DCRI as an analytical partner to accelerate data-driven approaches in drug development.

DCRI NEWS