Automatic determination of MS lesion subtypes based on fractal analysis in brain MR images

Abstract

In this paper a novel approach based on fractal analysis has been proposed to determine MS lesions into two subtypes (i.e., Enhancing lesions (Acute), T1 “black holes” (chronic) lesions) in Fluid Attenuated Inversion Recovery (FLAIR) MR images, automatically. In the proposed method, firstly, MS lesion voxels are segmented in FLAIR images using Entropy-Based EM Algorithm and Markov Random Field (MRF) model. Then, Fractal dimension of each lesion voxel is computed in FLAIR images and used with signal intensity features (T1-weighted, gadolinium enhanced T1-weighted, T2-weighted). Finally, a neural network classifier is applied to feature vectors. Evaluation of the proposed method was performed by manual segmentation of chronic and acute lesions in gadolinium enhanced T1-weighted (Gad-E-T1-w) images by studying T1-weighted (T1-w) and T2-weighted (T2-w) images, using similarity criteria. The results showed a good correlation between the lesions segmented by the proposed method and by experts manually. Thus, the suggested method is useful to reduce the need for paramagnetic materials in contrast enhanced MR imaging which is a routine procedure for separation of acute and chronic lesions.