MultiGeoMed: Multimodal Geometric Deep Learning in Medical Image Analysis

MultiGeoMed is a research initiative focused on advancing the use of geometric deep learning in medical imaging. The project seeks to address the following key questions:

1. Which geometric representation of medical volumes offers the greatest benefit for various downstream tasks?
2. Can multiple geometric representations be effectively combined to enhance the performance of geometric deep learning methods?
3. What are the most effective strategies for integrating these representations?

Geometric Representations

To explore these questions, the project will investigate four distinct geometric representations:

• Voxel grids
• Surface meshes
• Point clouds
• Implicit representations

Research Objectives

We aim to develop novel multimodal approaches for the following tasks:

• Shape analysis and completion
• Image registration
• Medical image classification

The project is funded by National Science Centre, Poland under SONATA Grant no: 2024/55/D/ST6/02081. It is hosted at the Sano Science Research Center.

Project Webiste: Link