UNet model for skull stripping and brain masks of anatomical images from PRIME-DE
Authors
Xindi Wang, Ting Xu
Description
The preprocessed brain masks of T1w images for all macaque monkeys from PRIME-DE. A convolutional network - UNet model was used to generate the brain mask for T1w images. The UNet model was initially trained in a large human sample and upgraded with a few macaque data. With a small macaque training sample (N=1-2), the UNet model achieves a decent performance of brain extraction with a minimal processing time (GPU: ~20s, CPU: 2-10 min).
Resource info table