Skip to content
View reubebe's full-sized avatar
  • DePauw University
  • Indianapolis
  • 17:52 (UTC -12:00)

Block or report reubebe

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
reubebe/README.md

REUBEN NEWTON ADDISON, Ph.D.

Clinical Research Scientist · Quantitative Scientist · Biomedical Data Science

clinical research · Bayesian inference · biomedical AI · movement science · reproducible analytics

Portfolio LinkedIn Google Scholar


About

I work at the intersection of clinical research, quantitative science, and machine learning. My background began in motor behavior and neurological movement disorders, then expanded into Bayesian modeling, biomedical data science, deep learning, and clinical trial design.

I am an Assistant Professor of Kinesiology and Director of the Motor Control Lab at DePauw University, and I work with the Chinese American Biopharmaceutical Society on data-science problems in biopharma. I also completed the Computational Data Analytics Track (Data Science) of Georgia Tech's M.S. in Analytics.

What interests me most is not building a model simply because the data allow it. I care about whether the analysis is reproducible, whether its assumptions are defensible, whether uncertainty is visible, and whether the result can support an actual scientific or clinical decision.

Current Focus

Bayesian clinical trial design · Biomedical machine learning · Parkinson's disease and movement science · ECG, EEG, motion, and longitudinal data · Reproducible evidence pipelines

Featured Projects

🧪 OpenTrial · Bayesian Clinical Trial Design Engine

OpenTrial turns structured trial inputs and external evidence into a cited, reproducible study-design workflow. It brings together evidence from ClinicalTrials.gov, PubMed, openFDA, and DailyMed, then uses that evidence to support Bayesian trial planning.

Key components include evidence-derived priors with source-level provenance, power and assurance analysis, prior-sensitivity analysis, predictive probability of success, and simulation-based group-sequential designs.

Python Streamlit Bayesian inference Simulation Clinical data APIs

Explore OpenTrial →


🫀 ECG Signal Quality Classification

A reproducible pipeline for identifying degraded cardiac signals in the PTB-XL clinical benchmark. The work combines conventional statistical baselines with deep learning and interpretability rather than treating the neural network as a black box.

The pipeline includes a CNN-Transformer hybrid ensemble, L1-regularized logistic regression, MLP and CNN baselines, stratified validation, threshold optimization, saliency analysis, and per-lead feature investigation.

Python PyTorch Signal processing Transformers Interpretability

Explore the project →


🦿 KinetiScan · Markerless 3D Motion Analysis

An ongoing computer-vision project for recovering metric 3D human movement from ordinary video. The system uses distance-conditioned modeling to lift 2D landmarks into 3D space and is being developed with clinical and biomechanical validation in mind.

Python PyTorch ONNX Computer vision Biomechanics

Technical Stack

Python R SQL MATLAB PyTorch scikit-learn Git LaTeX

Area Methods and tools
Machine Learning CNNs, Transformers, RNN/LSTM, MLPs, XGBoost, classification, regression, clustering
Bayesian & Statistical Modeling Bayesian inference, PyMC, mixed-effects models, longitudinal analysis, repeated measures, treatment-effect modeling
Clinical & Research Design Clinical trial design, experimental design, observational studies, sensitivity analysis, statistical analysis planning
Biomedical Data ECG, EEG, fMRI, TMS, motion capture, clinical trial evidence, behavioral and physiological time series
Reproducible Computing Python, R, SQL, MATLAB, Git/GitHub, Pandas, NumPy, scikit-learn, PyTorch

Current Work

At DePauw, I lead the Motor Control Lab, where my research centers on human movement, neurological function, and clinically meaningful measurement. My broader technical work now extends into biomedical AI and biopharmaceutical research.

Current directions include:

  • Bayesian and mixed-effects models for Parkinson's disease research
  • Longitudinal treatment-response and symptom-trajectory modeling
  • Reproducible pipelines for behavioral and neurophysiological data
  • Machine learning for biomedical and clinical datasets
  • Clinical trial design and evidence synthesis
  • Markerless motion analysis and computer vision
  • Teaching biomechanics, motor control, sports analytics, and applied data analysis

Selected Research

My research spans motor behavior, Parkinson's disease, neurophysiology, postural control, clinical technology, and remote healthcare delivery.

Selected work

  • Addison, R. N., & Van Gemmert, A. W. (2023). Bilateral transfer of a visuomotor task in different workspace configurations. Journal of Motor Behavior. DOI
  • Addison, R. N., et al. (2026). A Parsimonious Predictive Model for Hypertension in African American Adults in the Jackson Heart Study. American Journal of Hypertension. Under Review. Manuscript AJH-D-26-00235.
  • Hondzinski, J., et al., including Addison, R. N. (2026). Evidence for preserved gait consistency after wild blueberry ingestion by individuals with Parkinson's disease. Submitted to the Journal of Parkinson's Disease.
  • Gauss, T. M., Addison, R. N., et al. (2025). Wild blueberry supplementation and motor symptoms in Parkinson's disease. Society for Neuroscience.
  • Addison, R. N., Steinberg, F., & Van Gemmert, A. W. (2024). Neural activity associated with visuomotor adaptation in various workspace locations. NASPSPA.
  • Wang, Y., Hu, D., Addison, R. N., et al. (2023). Functional changes in superior temporal gyrus in focal dystonia. 6th International Dystonia Symposium.
  • Smith, A., Addison, R., et al. (2018). Remote mentoring using telemedicine platforms. Journal of Ultrasound in Medicine.

View my Google Scholar profile →

Education

Institution Degree
Georgia Institute of Technology M.S. in Analytics, Computational Data Analytics Track (Data Science)
Louisiana State University Ph.D. Kinesiology, Motor Behavior
Memorial University of Newfoundland M.S. Kinesiology
University of Ghana B.S. Psychology

My doctoral work focused on visuomotor adaptation and workspace manipulation, combining behavioral, cognitive, and neurophysiological approaches to motor learning.

Research & Professional Interests

Clinical trial design · Biomedical AI · Bayesian modeling · Precision medicine · Interpretable machine learning · Movement science · Signal processing · Longitudinal data · Reproducible research

I am especially interested in collaborations where rigorous quantitative work can improve how clinical or biomedical evidence is generated, interpreted, and used.

Connect

Portfolio · LinkedIn · Google Scholar · GitHub

📧 reuben.addison@gmail.com


Clinical research × quantitative science × machine learning

Popular repositories Loading

  1. DDP_FInal-assignment DDP_FInal-assignment Public

    Prediction of flower characteristics

    R

  2. Datascience-capstone Datascience-capstone Public

    Data science specialization Coursera

    R

  3. BMC BMC Public

    Forked from BMClab/BMC

    Notes on Scientific Computing for Biomechanics and Motor Control

    Jupyter Notebook

  4. dl-colab-notebooks dl-colab-notebooks Public

    Forked from tugstugi/dl-colab-notebooks

    Try out deep learning models online on Google Colab

    Jupyter Notebook

  5. the-incredible-pytorch the-incredible-pytorch Public

    Forked from ritchieng/the-incredible-pytorch

    The Incredible PyTorch: a curated list of tutorials, papers, projects, communities and more relating to PyTorch.

  6. my_todo-app my_todo-app Public

    This is my beginner project

    Python