Self-organizing maps for brain tractography in MRI

Abstract

Brain white matter fibers can be mapped synthesizing the diffusion tensors obtained from diffusion weighted magnetic resonance images. An important drawback in the determination of the fiber paths for tractography purposes occurs in uncertainty regions where at least two fiber paths crossover. In this study we proposed an artificial neural network approach to clarify the fiber tracts in these uncertainty regions. The implementation of the proposed method, called SOFMAT, leads to an optimal track converging to the underlying path as a result of self-organization of an artificial neural network. The promising results on the well-known artificial dataset called PISTE served to identify an affective network configuration, tested for various noise levels. Finally, the resulting tractography for human brain MR images are illustrated.