mritc: A Package for MRI Tissue Classification

Dai Feng, Luke Tierney

Main Article Content

Abstract

This paper presents an R package for magnetic resonance imaging (MRI) tissue classification. The methods include using normal mixture models, hidden Markov normal mixture models, and a higher resolution hidden Markov normal mixture model fitted by various optimization algorithms and by a Bayesian Markov chain Monte Carlo (MCMC) method. Functions to obtain initial values of parameters of normal mixture models and spatial parameters are provided. Supported input formats are ANALYZE, NIfTI, and a raw byte format. The function slices3d in misc3d is used for visualizing data and results. Various performance evaluation indices are provided to evaluate classification results. To improve performance, table lookup methods are used in several places, and vectorized computation taking advantage of conditional independence properties are used. Some computations are performed by C code, and OpenMP is used to parallelize key loops in the C code.

Article Details

Article Sidebar