Fuzzy-aided tractography performance estimation applied to brain magnetic resonance imaging

A recursive fuzzy inference system that can be applied to estimate the error probability of tracking algorithms used in medical image processing systems is proposed. Specifically, we are interested in the fiber bundles estimation process (fiber tracking) in diffusion tensor (DT) fields acquired via magnetic resonance imaging (MRI). As tracking algorithm we consider a previously developed probabilistic tracking algorithm (PTA). This paper studies the analogies between this tracking approach and a typical Multiple Hypotheses Tracing (MHT) system, which is closely related to fuzzy systems. This comparison leads to the development of a SAM (Standard Additive Model) fuzzy system that on-line provides the uncertainty of the decisions about the estimated fiber tracts. Experiments on both simulated and real DT-MR images demonstrate the validity of the method.