Interpreting Surface Curvature Data on 3D Bone Meshes

Project Overview

The goal of this project was to accurately measure surface curvature values of 3D meshes created from scanned human bones. Previously, the measurement of “robusticity” relied on subjective visual assessment, leading to imprecise values with limited analytical value.

To address this, I developed a system using open-source software MeshLab for analyzing 3D triangular meshes. The challenge in trying to measure surface curvature of a mesh is that it technically does not have a surface. Meshes are created out of many triangles, joined at the vertices to neighboring triangles. To solve this I ended up measuring the change in slope from each triangle’s vertex, as it relates to its neighboring vertex. By combining the software’s .ply output with custom Python scripts, I was able to provide a comprehensive statistical analysis of the curvature data.

This was the first project I worked on as a student intern, and it taught me a lot. I went into this project completely new to the software and technology involved. A lot of research was involved, through reading published papers and discussion with the lab members. Another part that was interesting was working with non-CS people, and being primarily responsible for the creation of the software aspect, from start to finish.

Project Repository

Find the code and detailed information in the GitHub repository.