Fiber tracking based on unsupervised learning

The brain white matter can be mapped noninvasively by diffusion tensor magnetic resonance image analysis. An important drawback in the determination of the fiber paths for tractography purposes occurs in uncertainty regions where at least two fiber paths cross. This study proposes artificial neural network approach to clarify the fiber tracts in these uncertainty regions. After the implementation of the proposed method, the best match to original path is achieved as a result of training the network. The method is applied on synthetic simulated fiber representations with various noise levels. The application gives promising results, so that the method will be applied in real diffusion tensor brain MR images as future study.