Skull stripping of the mouse brain on MR images is a crucial step for rodent neuroimaging preprocessing. The traditional methods for this task are time-consuming. To solve the problem, we present a deep learning model, U-Net with Nonlocal Position-aware (NPA) block using domain knowledge transfer. The results demonstrated that our end-to-end method achieves high dice scores in several MR modalities with ultrafast processing speed which is two orders of magnitude faster than atlas-based methods. To conclude, our automatic skull stripping approach may provide an alternative to previous complex preprocessing pipelines for high-throughput rodent neuroimaging applications.
This abstract and the presentation materials are available to members only; a login is required.