Abstract
The semantic parsing of brain magnetic resonance images is an important step in many applications that require the segmentation of medical images. In this chapter, we describe the most commonly used approach for image segmentation, namely atlas-based segmentation. We discuss the main components of atlas-based segmentation, including atlas-to-image registration, label fusion, and atlas selection. In addition, we also review different brain atlases that are frequently used as anatomical priors in image segmentation of neuroanatomical structures.
Original language | English |
---|---|
Title of host publication | Medical Image Recognition, Segmentation and Parsing |
Subtitle of host publication | Machine Learning and Multiple Object Approaches |
Publisher | Elsevier |
Pages | 307-335 |
Number of pages | 29 |
ISBN (Electronic) | 9780128025819 |
ISBN (Print) | 9780128026762 |
DOIs | |
State | Published - 1 Jan 2015 |
Externally published | Yes |
Keywords
- Brain atlases
- MR imaging
- Machine learning
- Registration
- Segmentation