Research
Our faculty are consistently contributing to the national conversation through their research.
Faculty and students in the Division of Biostatistics & Health Data Science are active in many areas of methodological and collaborative research, and regularly publish in high-impact statistics, biostatistics, and bio-medical journals.
Bayesian analysis: James Hodges, Joseph Koopmeiners, Thomas Murray, Cavan Reilly, Lianne Siegel, Eric Lock, Lin Zhang
Causal inference: Jared Huling, Joseph Koopmeiners, Chap Le, Thomas Murray, Wei Pan, David Vock, Julian Wolfson, Tianzhong Yang
Clinical trials: Anne Eaton, Lynn Eberly, Erika Helgeson, James Hodges, Joseph Koopmeiners, Chap Le, Xianghua Luo, Andrew Mugglin, Thomas Murray, James Neaton, Cavan Reilly, Lianne Siegel, Kyle Rudser, David Vock
Statistical genetics & computational biology: Saonli Basu, Weihua Guan, Eric Lock, Wei Pan, Cavan Reilly, Sandra Safo, Tianzhong Yang, Lin Zhang
Screening & diagnostic testing: Lynn Eberly, Joseph Koopmeiners, Chap Le
Machine learning: Saonli Basu, Lynn Eberly, Erika Helgeson, Jared Huling, Eric Lock, Wei Pan, Ashley Petersen, Cavan Reilly, Sandra Safo, David Vock, Julian Wolfson
Medical imaging: Lynn Eberly, Mark Fiecas, Joseph Koopmeiners, Lin Zhang
Meta-analysis: James Hodges, Lianne Siegel, Tianzhong Yang
Spatial statistics: Mark Fiecas, James Hodges, Lin Zhang
Survival analysis: Anne Eaton, Chap Le, Xianghua Luo, Thomas Murray, Kyle Rudser, David Vock
We pride ourselves on getting students involved in research early and often. Masters students complete a research-oriented Plan B project in collaboration with a faculty adviser. PhD students are involved in research assistantships starting from their first semester, and typically identify a dissertation adviser during the second year of the program. Dissertations follow the 3-paper model, so our students often graduate with multiple peer-reviewed publications.
Recent Student Project and Dissertation Titles:
- Grace Lyden, “Policy-Relevant Causal Effect Estimators for Heterogeneous Treatments and Exposures”
- Chuyu Deng, “Innovative Methods for Treatment Effect Heterogeneity & Calibration”
- Yuan Zhang, “Modifications of Q-learning to Optimize Dynamic Treatment Regimes”
- Bin Guo, “Integrative Statistical Methods in Genomics and Neuroimaging”
- Shannon McKearnan, “Statistical Methods for Organ Transplant”
- Charles Cain, “Statistical Considerations for Clinical Trials Aiming to Identify Individualized Treatment Rules”
- Andrew DiLernia, “New Estimation and Inferential Methods for Functional Connectivity Analysis”
- Alex Knutson, “Integrating Summarized Imaging and Genomic Data with GWAS for Powerful Endophenotype Association Testing in Alzheimer’s Disease”
- Lianne Siegel, “Estimating the Reference Range from a Meta-Analysis using Aggregate or Individual Participant Data”
- Adam Kaplan, “Context-Driven Prior Distributions in Genome–Wide Association Studies, Medical Device Adaptive Clinical Trials, and Genetic Fine-Mapping”
Biostatistics working groups are intended for University of Minnesota students and faculty and focus on various areas of faculty research. This is an opportunity for collaboration, peer-to-peer mentoring, and networking between student cohorts. If you are interested in participating in any of these working groups, please contact the working group leads below.
Active Working Groups
Title: Advanced Parametric and Non-Parametric Bayesian Methods (AdBAYES)
Faculty Lead(s): Steffen Ventz, Thierry Chekouo Tekougang
Description: For BHDS students and postdocs eager to explore cutting-edge Bayesian techniques and methods at the forefront of recent statistical and computational advancements. No prior Bayesian background shall be required—however, the vision is to create a group of researchers with an interest in learning and applying these methods in their dissertation/research. For each meeting, a student / group of students will be assigned a paper to read and present on each of the topics above, followed by discussions/ Q&A. Over the semester(s), the WG will be structured into different components including parametric Bayesian methods, non-parametric Bayesian approaches, Bayesian theory and foundations and computational strategies for Bayesian inference.
Title: Biostats for High-Dimensional Data Working Group
Faculty Lead(s): Sandra Safo, Steffen Ventz
Description: The purpose of this working group is to discuss (bio)statistical methods for analyzing high-dimensional data (i.e., a large number of variables relative to the number of observations). During each meeting, one or two student members present on their own ongoing research or a journal article in this area. Sometimes we have a faculty from another university presenting work in this area. We spend the last meeting of the semester discussing a career development topic chosen by the group.
Title: Comparative Effectiveness and Research Synthesis (CERS)
Faculty Lead(s): Lianne Siegel
Description: The mission of our working group is to develop and implement statistical methods to aid in research synthesis and comparative effectiveness research and to foster collaboration among UMN students and faculty conducting research in these areas. Our members develop statistical methods across many areas including multivariate and network meta-analysis, meta-analysis of diagnostic tests, causal inference in meta-analysis, and meta-analysis of normative data from both observational studies and randomized clinical trials. We are also interested in bridging the gap between statistical theory and practice (i.e. translational biostatistics) by publishing peer-reviewed high impact manuscripts that translate recent methodological advances to a clinical and epidemiological audience. We meet regularly to share our current work and discuss journal articles. Our meetings will follow a hybrid format (in person and Zoom options) to facilitate collaboration with colleagues at other institutions but we encourage all local participants to attend in person when possible. All are welcome!
Title: Causal Inference Working Group
Faculty Lead(s): Jared Huling, Marquis Hou
Description: The purpose of this working group is to discuss recent advancements and ongoing research in the division related to causal inference. Topics to be discussed include SMARTs/adaptive intervention designs, federated learning, transportability and other topics related to causal inference.
Title: Collaborative Biostatistician Club (CBC)
Faculty Lead(s): Biyue, Daien Langworthy, Ashley Petersen
Description: This working group will focus on tackling statistical issues and practical challenges that biostatisticians often encounter in a collaborative environment. The group plans to help equip students with skills to help them succeed as collaborative biostatisticians. Meetings may include the presentation of case studies, discussion of articles related to collaborative biostatistics skills, hackathons to learn and practice a specific analysis or programming tool, and professional development panels/talks with topics selected by students.
Title: MERGE (Mobile & E-Health Research Group)
Faculty Lead(s): Julian Wolfson, Erjia Cui, Marquis Hou
Description: The MERGE working group is focused on the applications of statistics to mobile and electronic health data, including data collected from smartphones, wearable sensors, and administrative health databases (EHR). Methods for these data are diverse and developing, but include machine learning, causal inference, and functional data analysis. Group meetings are open to students at any stage of their program, and will include presentation of ongoing projects, discussion of research papers, and tutorials on relevant technologies and skills. Meetings will be followed by some friendly table tennis in the CCBR Ballroom!
Title: Minnesota Complex Innovative Design Research (M-CIDeR) Working Group
Faculty Lead(s): Joe Koopmeiners, David Vock
Description: This working group aims to foster a community for discussing current research areas in complex and innovative clinical trial design with an emphasis on the on-going dissertation/plan B papers of the participating students. Faculty (and students) will also share ideas for papers, which may be opportunities for newer students to get involved with research in this area. Faculty will also aim to provide insights into the on-going or recent clinical trials that they are coordinating, including exposure to statistical analysis plans and protocols, and other nuts and bolts of conducting trials.
Title: Spatio + Temporal Modeling Group
Faculty Lead(s): Caitlin Ward, Harrison Quick, Mark Fiecas
Description: The mission of this working group is to discuss spatial and temporal research being conducted in the division, including medical imaging analysis, disease mapping, time-series analysis, synthetic data generation, and epidemic modeling. Meetings will typically involve discussions of ongoing student or faculty research and are open to all students and faculty interested in this area of research. We will have social time before all meetings to foster community and collaboration.
Title: Statistical Genetics/Omics Journal Club
Faculty Lead(s): Eric Lock, Saonli Basu
Description: The focus of the journal club is to introduce and discuss methodological developments and applications for the analysis of molecular data (genomics/omics) in different scientific domains. We have generally had a scientific theme each semester, with discussions on recent research directions. Visit the Journal Club webpage for more information.
Title: Survival & Longitudinal Analysis Working Group (SLAWG)
Faculty Lead(s): Anne Eaton, Xianghua Luo
Description: We will meet to discuss biostatistical methods for time-to-event data and longitudinal data, including discussing papers, sharing ongoing research and attending and discussing webinars.
