Silvia P. Canelón

Silvia P. Canelón

Postdoctoral Research Scientist

University of Pennsylvania

About me

Hello, and welcome! I’m a postdoctoral research scientist in the Department of Biostatistics, Epidemiology, and Informatics at the University of Pennsylvania Perelman School of Medicine. My research interests include applications of biomedical informatics in the public and population health fields. I work in the Boland Lab on projects that develop novel data mining methods to extract pregnancy-related information from Electronic Health Records (EHR) and study the relationship between environment and disease. Learn more about my research interests in publications.

I enjoy using R to optimize my research workflow and have noticed it making guest appearances elsewhere in my life. I’m certified as an RStudio Tidyverse Instructor and am passionate about R education for the public good. Keep up with my R tinkering in posts and teaching in talks. Thanks for stopping by!


  • Reproductive & sexual health
  • Prenatal & perinatal health
  • Social determinants of health
  • Environmental exposures
  • Data science for good
  • R education


  • Tidyverse Instructor Certification, 2020


  • Certificate in Biomedical Informatics, 2019

    University of Pennsylvania

  • Ph.D. in Biomedical Engineering, 2018

    Purdue University

  • B.S. in Biomedical Engineering, 2012

    University of Minnesota



Postdoctoral Research Scientist

University of Pennsylvania

Oct 2018 – Present Philadelphia, PA

Developing tools to mine Electronic Health Record (EHR) data for population-level analyses with the purpose of understanding female fertility and infertility related conditions.

Designing machine learning methods and algorithms with the goal of contributing software to the biomedical informatics field and making it accessible in open platforms


Graduate Research Assistant

Purdue University

May 2012 – May 2018 Indianapolis, IN

Characterized the type I collagen matrix produced by osteoblasts in vitro to assess structural, biochemical, mechanical, and biological properties. Assays used include atomic force microscopy, nanoindentation, Fourier transform infrared spectroscopy, and quantitative gene expression analysis.

Investigated the effect of (1) an induced reduction in molecular crosslinking on type I collagen matrix properties and (2) mechanical loading via equibiaxial substrate strain to determine the impact of environmental factors.


Recent & Upcoming Talks

Revealing Room for Improvement in Accessibility within a Social Media Data Visualization Learning Community

Data visualization accessibility talk to share what we found after scraping alternative (alt) text from data viz shared on Twitter as part of the #TidyTuesday social project.

The Impact of Sickle Cell Status on Adverse Delivery Outcomes Using Electronic Health Record Data

Applied Clinical Research Informatics: Solving Real World Problems. Oral Presentations S17: March 23, 2021 11:30am-1pm ET

Writing Presentations in R

R-Ladies Seattle package demonstrations showing how to make beautiful slides with xaringan and how to deploy them.

Posts & Publications

Recent Posts

Deploying xaringan Slides: A Ten-Step GitHub Pages Workflow

A ten-step workflow for creating an HTML xaringan slide deck and deploying it to the web using GitHub Pages

Becoming certified as an RStudio Tidyverse Instructor

An overview of the RStudio Instructor certification process and collection of resources to support anyone on their certification journey.

Customizing Hugo Academic's Dark Mode with Help from Atom

Tutorial on how to customize the dark mode in Hugo’s Academic theme with help from the Atom text editor package Pigments.

Recent Publications

Individual-Level and Neighborhood-Level Risk Factors for Severe Maternal Morbidity

Study providing evidence that neighborhood-level risk factors are independent predictors of Severe Maternal Morbidity, providing further evidence that racial disparities in maternal outcomes are symptoms of historical and structural racism.

Not All C-sections Are the Same: Investigating Emergency vs. Elective C-section Deliveries as an Adverse Pregnancy Outcome

Publication and poster accepted for the 2021 Pacific Biocomputing Symposium. This study utilizes Electronic Health Record (EHR) data to assess the impact of pregnancy-specific maternal morbidity and patient-specific characteristics on experiencing an emergency admission at the time of delivery and its relationship to Cesarean section (C-section) deliveries

Development and Evaluation of MADDIE: Method to Acquire Delivery Date Information from Electronic Health Records

An R algorithm designed to extract delivery episode details from structured Electronic Health Record data.


Send me a note