Sander L, et al. - The researchers presented a novel method for brainstem volumetry by utilizing a fully-automated segmentation method based on multi-dimensional gated recurrent units (MD-GRU), which was a deep learning based semantic segmentation method that used a convolutional adaptation of gated recurrent units on. An accurate, highly reproducible, and robust segmentation in high control and patients with MS and Alzheimer's disease in 200 s/scan on an Nvidia GeForce GTX 1080 GPU was provided by the fully automated brainstem segmentation method MD-GRU. Furthermore, it portrayed the potential for application in large and longitudinal datasets.