For the past four summers, Rho has participated in the Industrial Math/Stat Modeling Workshop (IMSM) hosted by the National Science Foundation-supported Statistical and Applied Mathematical Sciences Institute (SAMSI). The workshop, a joint program of Duke University, the University of North Carolina at Chapel Hill, and North Carolina State University, exposes graduate students in mathematics, engineering, and statistics to real-world challenges arising in industrial and government research. SAMSI students break out into small teams and spend a week collaborating on projects brought to IMSM by corporate and government research teams.
The 2014 IMSM workshop wrapped up this week. Rho was one of five project presenters, along the Centers for Disease Control and Prevention, the US Army Corps of Engineers, the MIT Lincoln Laboratory, and SAS. Rho was represented at the workshop by Rho’s CEO Dr. Russ Helms, Rho’s Vice President of Federal Operations, Dr. Herman Mitchell, and statistical scientist Agustin Calatroni. Rho partnered with SAMSI researcher, Dr. Sanvesh Srivastava, to present students with the challenge of developing predictive models for allergen exposure and asthma using data from the National Health and Nutrition Examination Survey (NHANES) 2005-2006.
Agustin Calatroni (far left) and Herman Mitchell (far right) with students from the SAMSI program.
NHANES 2005-2006 was the first study of its kind to compile nationwide data on asthma, allergies, and home allergen exposures. As a result, for the first time in history, researchers have access to a nationally-representative sample of allergen sensitization, asthma, and home environment data. One goal of the IMSM project is to tap into the home environment data to predict exposure to specific allergens which may exacerbate asthma. At present, the only definitive tests for allergen exposure are through direct sampling methods like dust collection. Direct sampling methods are more accurate than self-reports from surveys, but they are also inconvenient (for both families and researchers), expensive, and time consuming. If a reliable predictive model for specific allergen exposures could be developed based on self-reported questionnaire data, it may reduce the need for direct sampling.
The analytical challenge for the IMSM students was twofold. First, the team had to figure out how to manage the massive data set, including how to cope with missing information. Second, the team had to develop statistically and mathematically rigorous algorithms that could accurately predict asthma/exposure from the other information provided in the dataset.
The challenges addressed by the SAMSI team mirror the challenges Rho researchers work on every day – complex data management, applied biostatistics, innovative problem solving. More importantly, the students were given the opportunity to work on a real world problem – expanding our understanding of asthma so we can develop better treatments.
Rho’s values are driven by our core purpose: to improve health, extend life, and enhance quality of life through corporate and research excellence. Participating in events like the IMSM give us a chance to share not only our knowledge, but also our core values with the next generation of researchers, statisticians, and scientists. Our aim is to instill our passion for conducting research, solving problems, and improving health in these future leaders.
If you are interested in statistical innovation, check out this video featuring Agustin Calatroni discussing graphic visualization of multivariate data.