Master of Science
Obesity currently affects 25% of Canadians and is strongly associated with many diseases including diabetes, cardiovascular disease, and cancer. However, only Intra-Abdominal Adipose Tissue (IAAT) is an important predictor of obesity associated disease and mortality. Magnetic Resonance Imaging (MRI) is an effective modality for imaging fat and methods have been developed to automate segmentation of IAAT. Currently existing techniques for automated segmentation in mice require acquisition using high magnetic field MRI equipment and use image acquisition techniques with low precision for fat quantification. We demonstrate a new fully automated technique for fat quantification in clinical strength mouse MRI through adaptation of an existing human fat quantification technique. We validated this method using images collected from mice in a 3 T clinical MRI against manual segmentation. Dice correlation coefficients revealed that 84% of voxels agreed for Subcutaneous Adipose Tissue (SAT) and 87% of voxels agreed for IAAT.
McCurdy, Colin M., "Fully Automated Segmentation and Quantification of Abdominal Adipose Tissue Compartments in Mouse MRI" (2014). Electronic Thesis and Dissertation Repository. 2366.