Mapping the double bonds in triglycerides

This study presents and validates a theoretical model for estimating the number of double bonds in triglyceride molecules using magnetic resonance imaging. The model enables reliable estimation of the number of double bonds from a small number of time points, as are typically acquired with chemical shift imaging. Prior knowledge from the US Department of Agriculture (USDA) is used to constrain the properties of triglyceride. Validation in oil phantoms shows agreement between the measured number of double bonds and USDA reference values (correlation 0.95, significance P=.0003, slope 0.95±0.31, intercept 0.08±1.24). Feasibility in a human subject was demonstrated using a long breath-hold (43 s) scan.