Atlas-Free Automatic Segmentation of Sheep Brain MRI.

Automated 3D brain segmentation methods have been shown to produce fast, reliable, and reproducible segmentations from magnetic resonance imaging (MRI) sequences for the anatomical structures of the human brain. Despite the extensive experimental research utility of large animal species such as the sheep, there is limited literature on the segmentation of their brains relative to that of humans. The availability of automatic segmentation algorithms for animal brain models can have significant impact for experimental explorations, such as treatment planning and studying brain injuries. The neuroanatomical similarities in size and structure between sheep and humans, plus their long lifespan and docility, make them an ideal animal model for investigating automatic segmentation methods.This work, for the first time, proposes an atlas-free fully automatic sheep brain segmentation tool that only requires structural MR images (T1-MPRAGE images) to segment the entire sheep brain in less than one minute. We trained a convolutional neural network (CNN) model - namely a four-layer U-Net - on data from eleven adult sheep brains (training and validation: 8 sheep, testing: 3 sheep), with a high overall Dice overlap score of 93.7%.Clinical relevance- Upon future validation on larger datasets, our atlas-free automatic segmentation tool can have clinical utility and contribute towards developing robust and fully automatic segmentation tools which could compete with atlas-based tools currently available.

Roadmap for an imaging and modelling paediatric study in rural NZ.

Our study methodology is motivated from three disparate needs: one, imaging studies have existed in silo and study organs but not across organ systems; two, there are gaps in our understanding of paediatric structure and function; three, lack of representative data in New Zealand. Our research aims to address these issues in part, through the combination of magnetic resonance imaging, advanced image processing algorithms and computational modelling. Our study demonstrated the need to take an organ-system approach and scan multiple organs on the same child. We have pilot tested an imaging protocol to be minimally disruptive to the children and demonstrated state-of-the-art image processing and personalized computational models using the imaging data. Our imaging protocol spans brain, lungs, heart, muscle, bones, abdominal and vascular systems. Our initial set of results demonstrated child-specific measurements on one dataset. This work is novel and interesting as we have run multiple computational physiology workflows to generate personalized computational models. Our proposed work is the first step towards achieving the integration of imaging and modelling improving our understanding of the human body in paediatric health and disease.