Professor Maurice A. Curtis 

Neuroscience, The University of Auckland, NZ

Maurice began his career as a radiographer before switching to neuroscience, completing a Master of Science and a Doctor of Philosophy degree at the University of Auckland. Maurice’s research work is focussed on understanding the earliest changes that occur in the brain and other anatomical regions that indicate that Parkinson’s disease is going to strike. The goal of the research is to have degenerative brain diseases detected and treated years before brain cell loss occurs. In this presentation Maurice will talk about his group’s research to reconstruct at the histological level, the anatomy of the human olfactory system where Parkinson’s and Alzheimer’s neurodegeneration begin in the brain. Maurice will also highlight recent work using machine learning to solve big challenges in the field of anatomy. Maurice is Professor of Neuroscience, Co-director of the Neurological Foundation Human Brain Bank, and a founding Director of the Medical Imaging Research Centre at the University of Auckland.

---

Visualizing the Human Olfactory Projection in 3D Using High Performance Computing

Visualizing in humans the 3D microanatomy of the axonal projections of olfactory sensory neurons from the olfactory epithelium to the olfactory bulb together with the ancillary structures of bone and vasculature necessitates a workflow for handling a great many sections, and the image registration will be computationally expensive. Here, we assembled a 3D reconstruction starting from a 7.45 cm3 en-bloc specimen that we extracted from an embalmed human cadaver. A complete series of 10 µm coronal sections was subjected to quadruple fluorescence histology and scanned in four channels.
 
Structures of interest were manually segmented in the scanned images. Convolutional neural networks were then trained for automatic segmentation of these structures. A high-performance computing solution was engineered to register the sections based on the fluorescence signal and 91 structures segmented. The 3D reconstruction offers several didactic capabilities. We arrived at an approximation of the number of olfactory sensory neurons in human: 2,690,273.