Meeting Banner
Abstract #3510

A comparison of three whole brain segmentation methods for in vivo manganese enhanced MRI in animal models of Alzheimer’s disease

Igor Varga1,2, Hae Sol Moon3, Adam Conrad4, Matthew Holbrook3, Andrei R Niculescu5, Abinaya Lakshmanan3, Robert J Anderson5, Christina L Williams6, Cristian T Badea3,5, Po-Wah So2, and Alexandra Badea3,5,7,8
1Department of Cybernetics, Czech Technical University, Prague, Czech Republic, 2Department of Neuroimaging, King's College London, London, United Kingdom, 3Biomedical Engineering, Duke University, Durham, NC, United States, 4Georgia Institute of Technology, Atlanta, GA, United States, 5Radiology, Duke University Medical Center, Durham, NC, United States, 6Psychology and Neuroscience, Duke University, Durham, NC, United States, 7Neurology, Duke University, Durham, NC, United States, 8Brain Imaging and Analysis, Duke University Medical School, Durham, NC, United States

Whole mouse brain segmentation is an essential prerequisite for multiple quantitative image analysis tasks and pipelines. In this work, we compare three methods for whole brain segmentation (skull stripping) relying on active contours, graph cuts and convolutional neural networks. We applied these methods on mouse brain manganese enhanced MR images acquired at 100 micrometre isotropic resolution, in vivo. All three methods achieved Dice coefficients larger than 94%, but convolutional neural networks achieved a small but significant improvement (0.97±0.01) over our active contours implementation (0.94±0.05) and the difference approached significance relative to graph cuts (0.96±0.01).

This abstract and the presentation materials are available to members only; a login is required.

Join Here