A GA-based approach for parameter estimation in DT-MRI tracking algorithms

This paper expands upon previous work of the authors in the field of fiber tracking in diffusion tensor (DT) fields acquired via magnetic resonance (MR) imaging. Specifically, we now focus on tuning-up a previously developed probabilistic tracking algorithm by making use of a novel genetic algorithm which helps to optimize most of the adjustable parameters of the tracking algorithm. Since the adjustment of these parameters constitutes a hard NP-complete problem, traditionally, this task has been heuristically approached. A year ago, we presented in the WCE'07 a multilayer neural network that was successfully applied to this issue. Its robustness and complexity were studied with more detail in the extended version recently published in the IAENG journal on Computer Science. Sine complexity constituted its main drawback, in this paper we explore the possibility of using a computationally simpler method based on a microgenetic algorithm. This strategy is shown to outperform the NN-based scheme, leading to more robust, efficient and human independent tracking schemes.