Welcome

Hi, it’s nice to meet you! I’m Feng, a sixth-year MD-PhD student in Biostatistics. I completed my PhD in the PennSIVE group at the Perelman School of Medicine with Dr. Russell (Taki) Shinohara. My research interests include developing novel statistical and machine-learning-based methodology for neuroimaging and other big data settings. In neuroimaging, I’ve developed tools for multi-modal image analysis, unsupervised MRI segmentation problems, and image harmonization as well as authored a comprehensive review of current image harmonization methods. Statistically, I’ve proposed a privacy-preserving algorithm for distributed fitting of population-based reference charts. As I continue to grow as a biostatistician, I’m particularly interested in leveraging powerful and flexible methods, such as deep learning, in statistically-informed ways and building new methods alongside the scientists who will use them.

Outside of my methodological work, I’m interested in collaborating with biomedical researchers and clinicians to enable them to get the most out of their data in a statistically-sound way, as well as how to interpret studies and clinical trials from a statistical lens. To do so, I’ve served as a statistical collaborator for a number of biomedical, clinical, and health informatics projects. To formalize the process, organize real-life consulting experience for fellow Biostatistics students, and provide an accessible statistical service to fellow biomedical trainees, I founded the Biostatistical Student Analysis Core (BSAC) within the Biostatistics graduate program. If you’re a Penn biomedical trainee of any kind and are interested in working together, please reach out to us through our BSAC Intake Application! We’d love to hear from you!

Clinically, I’m interested in diagnostic radiology. I love the mix of diagnosis, medical thinking, and opportunity for procedure within this specialty. In this context, I look forward to the opportunity for developing new methods in the context of spatially-dependent imaging data, supported by clinical knowledge and a dedication for advancing patient care. Over the last few years of medical school and beyond, I’m excited to learn more!

View my CV here or at the “My CV” tab above!

Interests
  • Neuroimaging
  • Deep learning
  • Normative modeling
  • Federated/distributed learning
  • Clinical trials
  • Observational studies
  • scRNA-seq
  • Spatial statistics
Education
  • MD-PhD student, 2019—present

    University of Pennsylvania

  • PhD in Biostatistics, 2021—2024

    University of Pennsylvania

  • BA in Statistics and Biology, 2015—2019

    Amherst College

  • High school, 2012—2015

    Illinois Mathematics and Science Academy (IMSA)

Skills

R
Python
Zotero
Statistics
Neuroimaging
Cell biology

Research Experience

 
 
 
 
 
PhD
PennSIVE, Perelman School of Medicine
Jun 2020 – Sep 2024 Philadelphia, PA
  • Learn, apply, and extend theoretical statistics concepts in the biomedical setting.
  • Develop and validate statistically-motivated methodology for computational neuroscience.
  • Collaborate with neuroimaging scientists on applied segmentation and analysis problems.
 
 
 
 
 
BSAC founder, President, and consultant
BSAC
Aug 2019 – Present Philadelphia, PA
  • Work with Graduate Group leadership to build, organize, and run BSAC within PhD program time and funding constraints.
  • Lead outreach for new collaborations and connect collaborators to consultants with appropriate skillset.
  • Advocate for fellow BSAC consultants to get appropriate credit and conduct consultations that are limited in scope.
  • Provide statistical guidance and conduct analyses for basic science, clinical, and health informatics collaborators.
  • Draft methodological descriptions and create figures for publication.
 
 
 
 
 
Research assistant
Jeeyon Jeong Lab, Amherst College
Aug 2017 – Jun 2019 Amherst, MA
  • Investigate essential metal characteristics in weedy rice species to determine the metal uptake mechanisms.
  • Establish Arabidopsis thaliana lines with mutations in essential iron-deficiency response proteins.
  • Provide statistical guidance for experimental design and analysis of complex data.
  • Create publication-quality figures and draft manuscript.
 
 
 
 
 
Research coordinator
Lynn M. Yee Lab, Northwestern University
Jun 2016 – Aug 2019 Chicago, IL
  • Perform mixed-methods, thematic analysis on a program designed to improve postpartum return to care.
  • Investigate themes in text-message-based communication correlated with return for postpartum care.
 
 
 
 
 
Statistics and Data Science Fellow
Statistics Department, Amherst College
Mar 2017 – Jun 2019 Amherst, MA
  • Tutor students in Introduction to Statistics, Intermediate Statistics, and Probability in lecture materials, homework, and R code.
  • Review class handouts and create solution sets R worksheets.
  • Serve as a statistical analyst for various on-campus academic and administrative departments and projects.
 
 
 
 
 
Research assistant and translation coordinator
No Pain Labor & Delivery (NPLD), Northwestern University
Jun 2013 – Jan 2019 Chicago, IL
  • Write and edit manuscript draft describing impact of NPLD on labor and delivery outcomes in China
  • Collect and analyze longitudinal data on changes in labor and delivery outcomes across multiple Chinese hospitals.
  • Mediate conversation between American obstetricians, obstetrical anesthesiologists, neonatologists, and nurses and their Chinese counterparts in the delivery rooms of target hospitals for one week each summer. The goals were to facilitate use of labor analgesia, decrease Cesarean rates, and provide better neonatal care.
  • Provide instruction and guidance to a team of mostly American high school, college, and medical school students who are semi-fluent to fluent in Chinese.
  • Maintain and distribute an updating list of commonly encountered medical terminology and their translations.

All Publications

Quickly discover relevant content by filtering publications.
(2023). Image harmonization - A review of statistical and deep learning methods for removing batch effects and evaluation metrics for effective harmonization. NeuroImage.

PDF Cite

(2022). Voxel-wise intermodal coupling analysis of two or more modalities using local covariance decomposition. Human Brain Mapping.

PDF Cite Project

(2022). Translatability Analysis of National Institutes of Health–Funded Biomedical Research That Applies Artificial Intelligence. JAMA Network Open.

PDF Cite Project

(2022). Associations of mindfulness, food security, and health literacy with diabetes-related perinatal health outcomes. American Journal of Obstetrics & Gynecology.

Cite

Contact