Skip to content

Latest commit

 

History

History
47 lines (30 loc) · 1.65 KB

File metadata and controls

47 lines (30 loc) · 1.65 KB

Back to Projects List

MICUDA - Mutual Information with Extensive Entropy and CUDA

Key Investigators

  • Luiz Otavio Murta Jr. (University of Sao Paulo)
  • Vinicius Pavanelli Vianna (University of Sao Paulo)

Project Description

MICUDA is a tool to register images using mutual information and generalized entropies (Tsallis and Renyi) with GPU acceleration.

Objective

  1. Improve the registration range of ITK and 3D Slicer using generalized entropies
  2. Provide a faster registration tool for realtime applications
  3. Move towards a fully automated tool for batch registration (e.g. medical atlas generation)

Approach and Plan

  1. Presentation on monday (5 minutes):
    • Show a little demo of fast registration using GPU
    • Talk about the problems found in the ITK Mattes metric how to solve them
    • Discuss some mathematical visualization of the metrics output and how to find local extrema that challenges registration with it

Progress and Next Steps

  1. Describe specific steps you have actually done.
  2. ...
  3. ...

Illustrations

Background and References

You can also check our preprint at arXiv for more details on it