James Matuk

  • Assistant Professor

Contributions to Public Health

  • Longitudinal studies are useful for understanding exposures and health outcomes through time. Parametric statistical approaches can fail to describe phenomena. To address shortcomings of standard approaches, I have developed Bayesian non-parametric models that can flexibly capture relationships between exposures and outcomes.  
    • Matuk, J., Bharath, K., Chkrebtii, O. and Kurtek, S. Bayesian Framework for Simultaneous Registration and Estimation of Noisy, Sparse and Fragmented Functional Data. Journal of the American Statistical Association. 2022. 117 (540): 1964 – 1980.
    • Matuk, J., Bharath, K., Chkrebtii, O., and Kurtek, S. Geometric Empirical Bayesian Model for Classification of Functional Data under Diverse Sampling Regimes. IEEE/CVF Conference of Computer Vision and Pattern Recognition. June 2021. 4429-4437.
  • Modern methods for capturing health related data can result in structures that are difficult to analyze, such as images, trees, and functional data.  Successful analysis of these data requires one to account for the non-Euclidean spaces of which these data are elements. From this perspective, I have developed approaches to analyze complex data.
    • Matuk, J., Kurtek, S. and Bharath, K. Topo-Geometric Analysis of Variability in Point Clouds using Persistence Landscapes. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2024. 46 (12): 11035 – 11046.
    • Matuk, J., Mohammed, S., Kurtek, S. and Bharath, K. Biomedical Applications of Geometric Functional Data Analysis. Handbook of Variational Methods for Nonlinear Geometric Data. Switzerland: Springer International Publishing.  2020.  675-701.

Google Scholar

Education

August 2021, PhD, Statistics, The Ohio State University, Columbus, OH
December 2018, MS, Statistics, The Ohio State University, Columbus, OH
May 2016, BS, Mathematics, Duquesne University, Pittsburgh, PA

Department/Affiliation