Table B.1.

An overview of the methods and data used by the Challenge participants. We denote methods that are unsupervised with the letter U and those that require some training data (supervised methods) with the letter S.

Name	Approach	Sequences
	Multimodal patch matching with an l2-norm	T1-w, T2-w, PD-w, & FLAIR
	Robust EM initialized graph cut	T1-w, T2-w, & FLAIR
	Class specific sparse dictionaries	T1-w, T2-w, PD-w, & FLAIR
	Mixture of global & local intensity distributions from a reference population	T1-w, T2-w, & FLAIR
	n3 Convolutional Neural Networks	T1-w, T2-w, PD-w, & FLAIR
	Hierarchical MRF & random forest refinement	T1-w, T2-w, & FLAIR
	Random forests	T1-w, T2-w, PD-w, & FLAIR
	Hierarchical EM followed by temporal consistency check	T1-w& FLAIR
	n2 Convolutional Neural Networks	T1-w, T2-w, PD-w, & FLAIR
	Hierarchical subject specific GMM	T1-w, T2-w, & FLAIR