PigSNIPE: Scalable Neuroimaging Processing Engine for Minipig MRI

Translation of basic animal research to find effective methods diagnosing and treating human neurological disorders requires parallel analysis infrastructures. Small animals such as mice provide exploratory disease models. However, many interventions developed using small models fail translate use due physical or biological differences. Recently, large-animal minipigs have emerged in neuroscience both their brain similarity economic advantages. Medical image processing is a crucial part research, it allows researchers monitor experiments understand development. By pairing four reinforcement learning five deep UNet segmentation with existing algorithms, we PigSNIPE, pipeline for the automated handling, processing, analyzing large-scale data sets minipig MR images. PigSNIPE registration, AC-PC alignment, detection 19 anatomical landmarks, skull stripping, brainmask intracranial volume (DICE 0.98), tissue 0.82), caudate-putamen 0.8) under two minutes. To best our knowledge, this first tool aimed at large images, which can significantly reduce time resources needed neuroimages.