Department of Biostatistics

Yu Shyr’s primary research interests are in developing statistical and bioinformatic methodology, specifically in experimental design for next-generation sequencing data, adaptive clinical trials, and nonparametric smoothing methods for longitudinal binary data, as well as in collaborative research in the fields of cancer biology, cancer epidemiology, and clinical trials.

Gustavo Amorim centers around developing novel statistical methods for problems arising in public health studies. His aim is to develop semiparametric models that are flexible and robust to model misspecification. He is currently focusing on two-phase designs, measurement-error problems and ordinal data analysis.

Lynne Berry works with the software development team in the Vanderbilt Center for Quantitative Sciences, overseeing and directing individual projects as well as the overarching goals and priorities of the group.

Dan Byrne is an expert at artificial intelligence in healthcare.  He has more than 35 years of experience building AI tools and predictive models in healthcare, and then testing in rigorous randomized controlled trials whether they improve health outcomes.  His Cornelius team has created more than 25 AI and predictive tools, which are now being tested in the electronic health record.  Based on his experience and research he has specific knowledge and practical skills regarding how to successfully implement AI in healthcare.

 

Dr. Chandrasekhar’s research is motivated by real-world problems that need to be addressed with suitable statistical methods. She serves as a statistical collaborator for the TN Department of Children’s Services, Vanderbilt Center for Critical Illness, Brain Dysfunction and Survivorship, and CDCs Emerging Infections Program and is a co-I on multiple NIH funded studies. Her methodological research is mostly associated with challenges encountered in her collaborative work and include health disparity measures, provider profiling, area under the time curve and methods to address survivor/attrition bias. In addition, Dr. Chandrasekhar co-instructs the statistics course for Doctorate in Nursing Practice students at the School of Nursing.

Dr. Chen’s research lands on both novel statistical research and impactful biomedical research. Currently, her statistical research focuses on missing data, survival analysis (especially time-varying effect and treatment switching), Bayesian methods in big data, and penalized approach for 1<n/p<10. She also devotes time to develop statistical methods in using electronic health records (EHRs) and genetics information in precision medicine. Her other significant biomedical research areas include cancer, ophthalmology, infection disease, cardiovascular diseases, and suicide.

Dr. Choi has extensive experience analyzing big data, especially in fields of electronic health records (EHRs) databases and accelerometer data. Based on her expertise in pharmacokinetics (PK) and pharmacodynamics (PD) modeling, her current research focus is to construct a system for PK/PD/pharmacogenomics studies using EHRs, which would provide the foundation for dose-optimization support system for precision medicine. Along these lines, she has been developing medication dosing extraction algorithms from EHRs, which would also benefit medication-related studies in pharmacoepidemiology. In the research using accelerometer data, she developed a new algorithm for preprocessing massive accelerometer data, which has been widely used as it was implemented in an R package as well as commercial software ActiLife.

Mario Davidson is an Assistant Professor at the Vanderbilt University School of Medicine Department of Biostatistics.  He received his PhD from The Ohio State University (OSU) in “statistics education” whereby he studied quantitative, qualitative, and mixed methods research.  He is the lead biostatistician for medical education researchers and medical students.  His research interests are in medical education assessment and research and statistical collaboration. He serves as an analyst, consultant, mentor, and educator. He has assisted scores of students with posters, abstracts, and papers.  Dr. Davidson has been the director of the Introduction to Research with an Emphasis in Biostatistics and Statistics Collaboration courses at Vanderbilt University School of Medicine and an introductory statistics instructor at Columbus State Community College.   

As an Assistant in Biostatistics, Liping Du collaborates with basic science and clinical research scientists in different fields, focusing on writing statistical analysis plan for grant applications and helping them with experimental designs, statistical data analyses, results interpretation and manuscript preparation. She also works on clinical trial reporting for ClinicalTrials.gov for VICC, and does data analysis for epidemiology studies and high-dimensional proteomic/genomic/metabolomics research. Recently she starts to collaborate with Peabody at Vanderbilt on education-related statistical data analysis.

Shawn Garbett enjoys applying computational and numerical skills to modeling complex data.

Tebeb Gebretsadik is a biostatistician engaged in collaborative research including projects related to clinical asthma research (adult and pediatric), infectious disease/global health, and rheumatology/immunology.  She is interested in statistical methods in epidemiology, predictive modeling, analysis of electronic health records, multidimensional data, observational studies and causal inference studies. A lead research project focuses on the role of maternal and environmental influences on pediatric respiratory health with an interdisciplinary approach to translate into improved and comprehensive understanding of disease heterogeneity, treatment and prevention. 

Robert Greevy is currently the Director of the Biostatistics Graduate Program and Director of Health Services Research, Biostatistics. He teaches the first year graduate course 'Principles of Modern Biostatistics'. His statistical methods research interests include: restricted randomization methods and sample size adaptive randomization for clinical trials, inverse probability weighted and doubly robust estimators, sensitivity analyses for unmeasured confounding, and foundational tools for better statistical inference. His medical research interests include: comparative effectiveness of anti-hyperglycemic medications in an EHR derived cohort of 400 thousand veterans with type 2 diabetes, effects of acute kidney injury in an EHR derived cohort of 17 million veterans, and randomized controlled trials for smoking cessation and for diabetes management.

Dr. Hackstadt's research interests are in the development and application of statistical methods for the exploration and analysis of complex data, which often arise in large public health studies, observational studies and studies utilizing electronic health records. These statistical methods include multiple imputation, semi-parametric regression, causal inference tools, such as propensity scores and principal stratification, and Bayesian modeling.  Applications of interest include environmental biostatistics, gestational diabetes, and pharmacoepidemiology, particularly studies related to diabetes.

Dr. Harrell received his PhD in Biostatistics from UNC in 1979. Since 2003 he has been Professor of Biostatistics, Vanderbilt University School of Medicine, and was the department chairman from 2003-2017. He is Expert Statistial Advisor for the Office of Biostatistics for FDA CDER. He is Associate Editor of Statistics in Medicine, a member of the Scientific Advisory Board for Science Translational Medicine, a member of the Faculty of 1000 Medicine, and a member of the policy advisory board for the Journal of Clinical Epidemiology. He is a Fellow of the American Statistical Association and winner of the Association's WJ Dixon Award for Excellence in Statistical Consulting for 2014. He was the 2017 Visionary Speaker, Clinical Studies Coordinating Center, University of North Carolina Department of Biostatistics. His specialties are development of accurate prognostic and diagnostic models, model validation, clinical trials, observational clinical research, cardiovascular research, technology evaluation, pharmaceutical safety, Bayesian methods, quantifying predictive accuracy, missing data imputation, and statistical graphics and reporting.

Dr. Huang specializes in latent variable modeling, confirmatory factor analysis (CFA), structural equation modeling (SEM) and mediation analysis. He also has expertise in longitudinal data analysis, hierarchical linear models. He is proficient with Mplus statistical software.

Dr. Huang’s interests include the application of advanced statistical methodology for prevention research on substance use and HIV risk behavior. Dr. Huang has served on a number of NIH and CDC-funded studies focused on substance abuse and HIV prevention (totally approximately $10 million).

Bob Johnson is a part-time (semi-retired) faculty member whose primary role is teaching and serving on students’ graduate committees. He previously collaborated with primary care researchers with focus on cluster randomized trials and continues in an advisory role. His current research involves methods for estimating mortality measures—e.g. life expectancy—in small geographic areas.

Dr. Kang's research interests include spatio-temporal modeling, multiple testing problems, and their applications to imaging data, in particular, fMRI data. He is also interested in multi-modal imaging data analysis, e.g., fMRI and DTI.

Dr. Koyama's primary research interests revolve around flexible experimental designs for clinical trials and inference from the data that such flexible and adaptive designs produce, both in the Frequentist and Bayesian paradigms. He currently teaches the graduate-level course, "Clinical trials and experimental design," and his collaborative research interests include comparative effectiveness research of treatments for localized prostate cancer and association between Acute Lung Injury and air pollution, among others.

Christopher Lindsell is a collaborative biostatistician who is a leader in the application of rigorous of biostatistical methods in the acute care environment, and to the intersection between emergency care and public health. He has led data coordinating centers for numerous multi-center clinical trials, including FDA-regulated trials, and for epidemiological studies. He holds patents on risk stratification in septic shock, and he has contributed significantly to a number of NIH-funded networks, including the CTSA. His current focus is on learning health systems, leveraging clinical processes and data systems to enhance learning from pragmatic trials and observational studies, and for designing and implementing dissemination and implementation research.

Dr. Liu completed her doctoral degree in biostatistics at the University of Michigan and a postdoctoral fellowship at Fred Hutchinson Cancer Research Center, before joining the Department of Biostatistics as an Assistant Professor at Vanderbilt University. Dr. Liu serves as the faculty biostatistician for the Department of Emergency Medicine collaboration plan and the Vanderbilt Memory & Alzheimer’s Center. Her research interests are focused on biomarker evaluation, event history data analysis with application to cardiovascular disease, cancer and Alzheimer's disease. In addition, she is also working on developing statistical methods for risk prediction modeling and its practical implementation in electronical health records system.

Dr. Liu focuses on integrating multi-omics data to address fundamental biological questions and to study complex disease.

Dr. Nian is a collaborative biostatistician with special interests in causal inference for multilevel treatments, Bayesian adaptive design for clinical trials and provider profiling.

Lauren Samuels is a collaborative biostatistician who specializes in the design and analysis of observational studies in health services research. Her cohort selection app, Visual Pruner, is available at http://biostat.mc.vanderbilt.edu/VisualPruner.

Dr. Sheng is focusing on next generation sequencing data analysis, especially for RNAseq, DNAseq and miRNAseq data. He is also interested in bioinformatics in proteomics, glycomics and metabolimics, including algorithm development, data analysis and software implementation.

Dr. Shepherd's research involves developing and applying novel statistical methods to studies of HIV/AIDS and other diseases of global health importance.  His statistical research has focused on developing methods for causal inference, ordinal data analysis, and measurement error.

Dr. Shotwell is a collaborative statistician and biomedical researcher. His collaborative work includes topics in anesthesiology, geriatrics and quality aging, drug repurposing, and many others. Dr. Shotwell is also interested in the optimal design and statistical treatment of biomedical experiments, especially those incorporating "challenge-response" elements that are common, for example, in the study of pharmacokinetics and electrophysiology. Dr. Shotwell is active in mentoring graduate students, and has taught several courses offered by the Department of Biostatistics, including the 'Statistical Collaborations in Health Sciences' and 'Statistical Machine Learning' course.
 

Chris Slaughter focuses on flexible Bayesian regression models for longitudinal data. He works closely with a variety of collaborators in Gastroenterology, Pediatrics, and Neonatology to advance medical and basic science research.

Andrew Spieker is an Assistant Professor of Biostatistics at Vanderbilt University Medical Center. He received his Ph.D. in Biostatistics from the University of Washington prior to completing a postdoctoral fellowship at the University of Pennsylvania. His research interests are primarily centered on the integration of causal inference methodology into fields such as pharmacoepidemiologic association studies, cost-effectiveness research, and preliminary HIV vaccine trials. Other research interests include the proper design, conduct, analysis, and interpretation of clinical studies; methods to diagnose and monitor disease; and risk factor identification.

Dr. Stewart is a clinical educator and collaborative statistician.  He teaches in both MSCI and MPH programs.  He has collaborated with a wide variety of researchers around the medical center, with substantial support for Nephrology and AHSQC.  His area of statistical methods research is missing data, with a special focus on support vector machines.

Ran Tao is interested in developing novel statistical methods to solve problems arising in the design and analysis of modern biomedical and public health studies, including genome-wide association studies, next-generation sequencing studies, and electronic health records systems. His current research topics include two-phase designs, complex survey sampling, rare-variant analysis, and risk gene prediction.

Simon Vandekar's work develops theoretically justified methods and software that are directly applicable in another scientific field. He is interested in applications in psychiatry, psychology, neuroimaging, and ecology.

 

Dr. Yu’s methodological research focuses on models for correlated binary data, mixture models, disease clustering, measurement error correction models, and statistical methods for genetics. Some of his works are cited in Univariate Discrete Distributions by Johnson, Kemp, and Kotz (3rd edition, 2005). Before joining Vanderbilt, Dr. Yu worked more than 5 years on clinical trials at Merck Research Laboratories in Rahway, NJ. He collaborates with investigators in clinical pharmacology, epidemiology, pulmonary, and Vanderbilt Institute for Clinical and Translational Research. He leads the Bayesian adaptive clinical trial design effort in the department.

Dr. Zhao's research interest is in developing tools with statistical methods to solve problems in next generation sequencing data analysis. He has developed tools in Rna-Seq sample size estimation, DNA-Seq and RNA-Seq data quality control. He is also interested in different kinds of high-dimensional data researches, including transcriptomics, genomics, and proteomics data analysis and integrating.

Dr. Zhu has training in both medicine and biostatistics, with expertise in advanced modeling concepts, predictive analytics, and statistical methods related to infectious disease.  She has been involved in influenza surveillance since 1999 and is interested in using large state or national data such as the National Ambulatory Medical Care Survey (NAMCS), the National Hospital Ambulatory Medical Care Survey (NHAMCS), Healthcare Cost and Utilization Project (HCUP), and Hospital Discharge Data System (HDDS) data etc. to improve public health.  She enjoys teaching and provides consultation services to graduate public health students.  Her current projects include influenza vaccine effectiveness, disease burden analysis, and studies with EHR-based prognostic decision support.