Meeting Banner
Abstract #3915

Automated Multi-Atlas Segmentation of Suspected Brown Adipose Tissue from Water-Fat MRI: Initial Evaluation

Elin Lundström1, Robin Strand1,2, Anders Forslund3,4, Peter Bergsten5, Daniel Weghuber6,7, Matthias Meissnitzer8, Håkan Ahlström1, and Joel Kullberg1

1Department of Radiology, Uppsala University, Uppsala, Sweden, 2Department of Information Technology, Uppsala University, Uppsala, Sweden, 3Department of Women's and Children's Health, Uppsala University, Uppsala, Sweden, 4Children Obesity Clinic, Uppsala University Hospital, Uppsala, Sweden, 5Department of Medical Cell Biology, Uppsala University, Uppsala, Sweden, 6Department of Paediatrics, Division of Paediatric Gastroenterology, Hepatology and Nutrition, Paracelsus Medical University, Salzburg, Austria, 7Obesity Research Unit, Paracelsus Medical University, Salzburg, Austria, 8Department of Radiology, Paracelsus Medical University, Salzburg, Austria

Segmentation of brown adipose tissue (BAT) from water-fat MR images generally requires time-consuming manual delineation. In this work a fully automated method, based on multi-atlas registration, for segmentation of human cervical-supraclavicular adipose tissue (suspected BAT) was evaluated using a semi-automated reference method, based on manual delineation. The presented method shows promising results for automated segmentation that allows time-efficient and objective measurements of BAT in large cohort research studies.

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

Join Here