Scientific publications

The problem of segmentation of three-dimensional fMRI images based on the Bayesian approach is considered, whereMarkov Random Field is used as the prior distribution, and von Mises-Fisher distribution is used as the observation model.Themainproblemwhenapplyingthisapproachinpracticeisanestimationofthemodelparameters.Inthispaper,wereviewalgorithms HMRF-MCEM, HMRF-EM and GrabCut, which implement this statistical model and estimate parameterswithout the usage of the labeled training data. The methods HMRF-EM and GrabCut were introduced in conjunctionwith other statistical models, but after a small modification, they can be used with the von Mises-Fisher distribution. Acomparative study was carried out by performing experiments on both synthetic, generated from the statistical model, andreal fMRI data.