Given a set of labelled MR images (atlases) and unlabelled images (subjects), MAGeT produces a segmentation for each subject using a multi-atlas voting procedure based on a template library made up of images from the subject set.
The major difference is that, in MAGeT brain, segmentations from each atlas (typically manually delineated) are propogated via image registration to a subset of the subject images (known as the ‘template library’) before being propogated to each subject image and fused. It is our hypothesis that by propogating labels to a template library, we are able to make use of the neuroanatomical variability of the subjects in order to ‘fine tune’ each individual subject’s segmentation.
M Mallar Chakravarty, Patrick Steadman, Matthijs C van Eede, Rebecca D Calcott, Victoria Gu, Philip Shaw, Armin Raznahan, D Louis Collins, and Jason P Lerch. Performing label-fusion-based segmentation using multiple automatically generated templates. Hum Brain Mapp, 34(10):2635–54, October 2013. (doi:10.1002/hbm.22092)
Hippocampus and subfields:
Jon Pipitone, Min Tae M Park, Julie Winterburn, Tristram A Lett, Jason P Lerch, Jens C Pruessner, Martin Lepage, Aristotle N Voineskos, and M Mallar Chakravarty. Multi-atlas segmentation of the whole hippocampus and subfields using multiple automatically generated templates. Neuroimage, 101:494–512, November 2014. (doi:10.1016/j.neuroimage.2014.04.054)
Min Tae M Park, Jon Pipitone, Lawrence H Baer, Julie L Winterburn, Yashvi Shah, Sofia Chavez, Mark M Schira, Nancy J Lobaugh, Jason P Lerch, Aristotle N Voineskos, and M Mallar Chakravarty. Derivation of high-resolution MRI atlases of the human cerebellum at 3T and segmentation using multiple automatically generated templates. Neuroimage, 95:217–31, July 2014. (doi:10.1016/j.neuroimage.2014.03.037)