I am a postdoctoral researcher at the Onnela Lab in the Department of Biostatistics at Harvard T.H. Chan School of Public Health. I am working on methods for digital phenotyping and its applications with Beiwe platform.
My statistical methods interests include: methods for processing, features extraction and analysis of accelerometry data, power estimation in complex settings, functional regression methods, machine learning, R software development.
I received PhD in Biostatistics from Johns Hopkins Bloomberg School of Public Health in December 2021. I received my Bachelor’s and Master’s degrees in Mathematics from Wroclaw University of Science and Technology in Poland in 2013 and 2015. Prior to joining Johns Hopkins, I worked as a research assistant at Indiana University and as an analyst for Opera Software Internet browser company.
In my free time, I enjoy connecting with people and running. I completed my first full marathon in 2017 in Denver, CO.
Download my CV.
Our work “Estimating knee movement patterns of recreational runners across training sessions using multilevel functional regression models” is accepted. This work can serve as a reference for practitioners modeling repeated functional measures at different resolution levels in the context of biomechanics and sports science applications.
Our paper “Comparison of Accelerometry-Based Measures of Physical Activity: Retrospective Observational Data Analysis Study” is published. This work provides comparison and harmonization mapping between minute-level accelerometry-derived measures (ActiGraph AC, MIMS, ENMO, MAD, AI).
Our paper “Quantification of acceleration as activity counts in ActiGraph wearables” is out. This work publishes activity counts algorithm from ActiGraph’s ActiLife and CentrePoint.
I was elected as a member of the Alpha chapter of the Delta Omega Society.
See all past updates.
We evaluate the power and sample size estimation properties of the upstrap resampling method.
We propose adaptive empirical pattern transformation (ADEPT), a fast, scalable, and accurate method for pattern segmentation in time-series.
Package adeptdata was created to host raw accelerometry data sets and their derivatives.
Package runstats provides methods for fast computation of running sample statistics for time series via Fast Fourier Transform.
We propose riPEER regularization method to estimate association between the brain structure features and a scalar outcome in a regression model while utilizing additional information about structural connectivity between the brain regions.
See for past academic years.