This lecture introduces the problem of medical image segmentation and looks in depth at two classes of algorithms that represent part of the state of the art: the Bayesian approach, on the one hand, and decision trees/random forests, on the other. The third part of the lecture deals with the close links between registration and segmentation, and presents segmentation algorithms based on atlas registration.
This lecture is complemented by the following two seminars:
- "Measuring the digital brain", Jean-François Mangin (Neurospin Saclay);
- Reconstructing organs in shapes", Hervé Delingette (Inria, Sophia-Antipolis).
