Ihor Varha presents Bayesian approaches in multi-modal MRI segmentation (reading group)

Bayesian approaches in multi-modal MRI segmentation, reading group presented by
Igor Varga, K333 (varhaiho@fel.cvut.cz)

----

Statistical shape models provide a framework which is capable of combining
strong prior information (e.g. a 3D anatomical atlas) with patient data, such
as
multi-modal MRI. In this reading group, the main focus will be on application
of
this approach to sub-cortical MRI segmentation, represented mainly by Visser
et al. [1]. The paper presents a method which allows learning from unsupervised
data and combining multiple MRI modalities at the same time.

Preceding works by Patenaude et al. [2],[3],
are recommended for a well explained overview and introduction to the topic
through a simple single-modality supervised bayesian version of the method.

----

[1] Visser, E., Keuken, M. C., Douaud, G., Gaura, V., Bachoud-Levi, A. C.,
Remy,
P., … Jenkinson, M. (2016). Automatic segmentation of the striatum and globus
pallidus using MIST: Multimodal Image Segmentation Tool. NeuroImage, 125,
479–497. https://doi.org/10.1016/j.neuroimage.2015.10.013

[2] Patenaude, B., Smith, S. M., Kennedy, D. N., & Jenkinson, M. (2011). A
Bayesian model of shape and appearance for subcortical brain segmentation.
NeuroImage, 56(3), 907–922. https://doi.org/10.1016/j.neuroimage.2011.02.046

[3] Patenaude, B. (2007). Bayesian Statistical Models of Shape and Appearance
for Subcortical Brain Segmentation. Department of Clinical Neurology, PhD
doctoral thesis