Yei Eun Shin, New Tenure-track Investigator in Biostatistics
, by Justine E. Yu, Ph.D.
Yei Eun Shin, Ph.D., M.S., was appointed in Spring 2021 as a tenure-track investigator in the Biostatistics Branch (BB), where she will build a collaborative research program designing complex models to study the impact of spatiotemporal exposures, such as physical activity and light exposures on risk of cancer and other diseases, in epidemiological cohort studies.
Advances in computer technology and software over the last 20 years have aided the management of large datasets from cohort and case-control studies to integrate spatial and temporal associations with outcomes. However, characterization of multidimensional structures such as space and time cannot be easily defined by deterministic physical laws without uncertainties. Statistical methods can enhance these studies by characterizing multidimensional spatiotemporal information, such as population mobility, radiation dosimetry, environmental exposures, climate changes, molecular dynamics, and viral transmission. Throughout her career, Dr. Shin has developed statistical approaches to advance research on these fronts in order to answer important scientific questions.
Describing Disease Progression with Spatiotemporal Models
For her predoctoral work, Dr. Shin described the disease progression pattern of Amyotrophic Lateral Sclerosis (ALS), also known as Lou Gehrig’s disease, using an autologistic regression model. ALS is a neurological disease that mainly affects the nerve cells in the brain and the spinal cord that control voluntary muscle movements. The symptoms typically start in a particular group of muscles and spreads as the disease progresses. Dr. Shin’s proposed model characterized how the disease spreads over space and time, and addressed challenges related to statistical modeling and inference. “The typical model assumes correlation between nearby tissues, but by breaking that assumption, we can detect new correlations between distant tissues,” said Dr. Shin. These findings can help neuroscientists understand the physiology and disease progression of ALS. She is extending this work to more complex and diverse spreading patterns to reflect realistic disease progression.
Improving Estimates in Epidemiological Studies
As a research fellow, Dr. Shin has developed survival analysis methods for two-phase cohort studies that use survey calibration. Two-phase cohort designs reduce the cost and effort of epidemiological studies by obtaining exposure data for a selected subset (phase two) of a larger cohort (phase one). Dr. Shin has made strides in improving estimation of two-phase designs. “Typically to estimate risk, weighted approaches are applied to data from the selected subset, however, utilizing baseline data from the entire cohort such as demographics and survival outcome could improve risk measurements,” Dr. Shin said. To improve the precision of absolute and relative risk estimates, she developed a weight calibration method that used individual follow-up times along with exposure information. She is currently exploring this method to address other issues in risk modeling and etiologic research including multiplicative-additive hazards models, cumulative incidence regression models, and complicated two-phase design risk estimates.
Integrating Spatiotemporal Modeling in Two-Phase Studies
In her new role, Dr. Shin will combine her expertise in statistical methodology and survival analysis methods for two-phase cohort studies to study circadian rhythm analysis using wearable-device-measured data of national survey studies. She is developing semi-parametric approaches to investigate individual biorhythms such as sleep-wake cycles from physical activity and ambient light exposure. Her long-term goal is to develop contemporary statistical methodologies for analyzing the risk of health-related outcomes associated with complex circadian rhythms and spatiotemporal environmental exposures to contribute to the understanding of cancer epidemiology and to improve public health.
Dr. Shin earned a B.S. and M.S. in statistics from Seoul National University, Korea. She earned her Ph.D. in statistics from Texas A&M University in College Station. She joined DCEG in 2017 as a postdoctoral fellow in BB and was promoted to research fellow in 2019.