Data Studio Workshop: High-Dimensional Fluorescent Microscopy for Guiding Therapy in Ulcerative Colitis Patients

Investigator: Stephan Rogalla, Department of Medicine, Division of Gastroenterology & Hepatology
Investigator: Derek Holman, Department of Radiology, Molecular Imaging Program
Investigator: Aaron Mayer, Department of Radiology, Molecular Imaging Program

Summary:
The Data Studio Workshop brings together a biomedical investigator with a group of experts for an in-depth session to solicit advice about statistical and study design issues that arise while planning or conducting a research project. This week, the investigator(s) will discuss the following project with the group.

The unique characteristics of tissues arise from the identities of their constituent cells and how these cells interact and communicate. We use spatially-resolved high-dimensional fluorescent microscopy to uncover the spatial cellular organization of gut tissues from patients with Ulcerative Colitis. We are interested in using these tissue architectural themes to predict whether patients will respond to therapy. By representing our microscopy images as Voronoi tessellations and assigning fluorescence intensities to each polygon, we are able to readily extract both cell types and their spatial relationships. We then generate classiers using machine learning algorithms and logistic regression.

Questions:
Our statistical questions for this workshop focus on three areas: 1) sample size for validation cohorts and study design when eects and eect sizes are unknown (as for machine learning classiers), 2) statistical analysis for features that take spatial relationships into account, and 3) multi-scale feature extraction.