Deep Learning C2 Spinal Fracture Prediction

This project featured a collaboration with the Radiology department at the University of Vermont Medical Center. The goal of the project was to design an end-to-end deep learning systems for the automated prediction of C2 spinal fractures from CT scans.

The core pieces of this project were first localizing the C2 vertabrae from the larger image, and then training various binary classification CNN’s.

See the full project write-up here

This project, as a piece of the Generalized Diagnostic Framework, won the 2018 UVM CS fair.


  • section_layout: 2col images:
    • caption: Example C2 segmentation description: Figure 1 url: /projects/fractures/ex.png