Abstract: One direction in medical image analysis is to effectively represent knowledge and efficiently extract biomedical information (such as a deformable shape) from medical images.  In particular, machine learning techniques, such as boosting, support vector machine (SVM), and graphical models, have played increasingly important role. Boosting, often used to learn discriminative models, has the advantage of automatically selecting and fusing informative features, which is a very appealing property in analyzing medical data where the image pattern is rather weak. The goal of this tutorial is to provide a comprehensive assessment of discriminative learning techniques used for medical imaging applications such as anatomical structure detection and segmentation, image categorization, etc. Learning from an annotated dataset the covers the uncertainties involved in the applications, these techniques are able to derive compact descriptions between the image and knowledge and gain improvements in performance and speed when compared with conventional algorithms without using learning.

Topics:

• Basics of discriminative models for medical imaging
• Ensemble classifiers
• Marginal space learning for medical image analysis
• Learning-based brain image analysis
• Learning approaches for brain surface analysis