Overview

This project aims to develop a more biologically informed approach for registering 3D models of anatomical structures, such as mouse skulls. Traditional registration methods (e.g., Coherent Point Drift) treat point clouds in a purely geometric way and often ignore the underlying biological constraints of how real organisms vary in shape. By incorporating statistical shape models—built from actual biological data—into the registration pipeline, the team seeks to achieve more accurate correspondences between specimens. This improvement will ultimately enhance automated measurements, morphological analyses, and potentially benefit broader applications where 3D biological data play a role, such as comparative anatomy or medical imaging.

Scope of Project

The work is led by Dr. Arthur Porto, Curator of AI at the Florida Museum, and supported by PhD student Breanna Shi. The collaborating team, including several Georgia Tech OMSCS students, will begin by studying existing implementations of Coherent Point Drift, understanding the current limitations of naive geometric registration, and then modifying the underlying mathematical kernels to reflect real-world biological variance. The project will involve:

• Data Preparation: Utilizing CT-scanned or photogrammetry-derived 3D models of mouse skulls and corresponding manual landmarks.
• Algorithmic Development: Extending or replacing parts of the Coherent Point Drift algorithm to incorporate statistical shape models (e.g., using PCA-derived covariance structures).
• Validation and Comparison: Measuring performance by comparing automated registrations to expert-defined landmark correspondences.
• Deliverable and Future Impact: Creating a user-friendly tool or workflow that biologists and other researchers can adopt for accurate, large-scale 3D shape analysis.

Collaboration and communication will occur regularly to ensure consistent progress. The broader vision is to produce software and methods that can handle the complexity of biological shapes at scale, fostering new insights in evolutionary biology, genetics, and beyond.

Meeting the Team

*Placeholder for Team Pictures/ Introductions*