Analysis of Generalized Entropies in Mutual Information Medical Image Registration
This work addresses robustness and speed issues in automated medical image registration, which is incremental as it builds on existing MI methods with GPU optimizations.
The study compared the performance of standard mutual information (MI) registration methods with custom GPU-accelerated Shannon and Tsallis MI functions under various transforms in 3D medical image registration, finding that the custom algorithms improved registration quality and achieved significant speed gains.
Mutual information (MI) is the standard method used in image registration and the most studied one but can diverge and produce wrong results when used in an automated manner. In this study we compared the results of the ITK Mattes MI function, used in 3D Slicer and ITK derived software solutions, and our own MICUDA Shannon and Tsallis MI functions under the translation, rotation and scale transforms in a 3D mathematical space. This comparison allows to understand why registration fails in some circumstances and how to produce a more robust automated algorithm to register medical images. Since our algorithms were designed to use GPU computations we also have a huge gain in speed while improving the quality of registration.